Pulling together some cleanups, fixes, and prepatory tci stuff. Most of this has been reviewed, but not all. Those lacking review: 01-tcg-aarch64-Fix-constant-subtraction-in-tcg_out_adds.patch 02-tcg-aarch64-Fix-I3617_CMLE0.patch 03-tcg-aarch64-Fix-generation-of-scalar-vector-operatio.patch 04-tcg-tci-Use-exec-cpu_ldst.h-interfaces.patch 06-tcg-Manage-splitwx-in-tc_ptr_to_region_tree-by-hand.patch 23-accel-tcg-rename-tb_lookup__cpu_state-and-hoist-stat.patch 24-accel-tcg-move-CF_CLUSTER-calculation-to-curr_cflags.patch 25-accel-tcg-drop-the-use-of-CF_HASH_MASK-and-rename-pa.patch 26-include-exec-lightly-re-arrange-TranslationBlock.patch 27-accel-tcg-Precompute-curr_cflags-into-cpu-tcg_cflags.patch Alex, the last patch is a re-write and extension of one that you did review. r~ Alex Bennée (4): accel/tcg: rename tb_lookup__cpu_state and hoist state extraction accel/tcg: move CF_CLUSTER calculation to curr_cflags accel/tcg: drop the use of CF_HASH_MASK and rename params include/exec: lightly re-arrange TranslationBlock Richard Henderson (23): tcg/aarch64: Fix constant subtraction in tcg_out_addsub2 tcg/aarch64: Fix I3617_CMLE0 tcg/aarch64: Fix generation of "scalar" vector operations tcg/tci: Use exec/cpu_ldst.h interfaces tcg: Split out tcg_raise_tb_overflow tcg: Manage splitwx in tc_ptr_to_region_tree by hand tcg/tci: Merge identical cases in generation (arithmetic opcodes) tcg/tci: Merge identical cases in generation (exchange opcodes) tcg/tci: Merge identical cases in generation (deposit opcode) tcg/tci: Merge identical cases in generation (conditional opcodes) tcg/tci: Merge identical cases in generation (load/store opcodes) tcg/tci: Remove tci_read_r8 tcg/tci: Remove tci_read_r8s tcg/tci: Remove tci_read_r16 tcg/tci: Remove tci_read_r16s tcg/tci: Remove tci_read_r32 tcg/tci: Remove tci_read_r32s tcg/tci: Reduce use of tci_read_r64 tcg/tci: Merge basic arithmetic operations tcg/tci: Merge extension operations tcg/tci: Merge bswap operations tcg/tci: Merge mov, not and neg operations accel/tcg: Precompute curr_cflags into cpu->tcg_cflags accel/tcg/tcg-accel-ops.h | 1 + include/exec/exec-all.h | 19 +- include/exec/tb-lookup.h | 26 +- include/hw/core/cpu.h | 2 + accel/tcg/cpu-exec.c | 34 ++- accel/tcg/tcg-accel-ops-mttcg.c | 3 +- accel/tcg/tcg-accel-ops-rr.c | 2 +- accel/tcg/tcg-accel-ops.c | 8 + accel/tcg/tcg-runtime.c | 6 +- accel/tcg/translate-all.c | 18 +- linux-user/main.c | 1 + linux-user/sh4/signal.c | 8 +- linux-user/syscall.c | 18 +- softmmu/physmem.c | 2 +- tcg/tcg.c | 29 +- tcg/tci.c | 526 ++++++++++---------------------- tcg/aarch64/tcg-target.c.inc | 229 +++++++++++--- tcg/tci/tcg-target.c.inc | 204 +++++-------- 18 files changed, 526 insertions(+), 610 deletions(-) -- 2.25.1 An hppa guest executing 0x000000000000e05c: ldil L%10000,r4 0x000000000000e060: ldo 0(r4),r4 0x000000000000e064: sub r3,r4,sp produces ---- 000000000000e064 000000000000e068 sub2_i32 tmp0,tmp4,r3,$0x1,$0x10000,$0x0 after folding and constant propagation. Then we hit tcg-target.c.inc:640: tcg_out_insn_3401: Assertion `aimm <= 0xfff' failed. because aimm is in fact -16, but unsigned. The ((bl < 0) ^ sub) condition which negates bl is incorrect and will always lead to this abort. If the constant is positive, sub will make it negative; if the constant is negative, sub will keep it negative. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/aarch64/tcg-target.c.inc | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/tcg/aarch64/tcg-target.c.inc b/tcg/aarch64/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/aarch64/tcg-target.c.inc +++ b/tcg/aarch64/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void tcg_out_addsubi(TCGContext *s, int ext, TCGReg rd, } } -static inline void tcg_out_addsub2(TCGContext *s, TCGType ext, TCGReg rl, - TCGReg rh, TCGReg al, TCGReg ah, - tcg_target_long bl, tcg_target_long bh, - bool const_bl, bool const_bh, bool sub) +static void tcg_out_addsub2(TCGContext *s, TCGType ext, TCGReg rl, + TCGReg rh, TCGReg al, TCGReg ah, + tcg_target_long bl, tcg_target_long bh, + bool const_bl, bool const_bh, bool sub) { TCGReg orig_rl = rl; AArch64Insn insn; @@ -XXX,XX +XXX,XX @@ static inline void tcg_out_addsub2(TCGContext *s, TCGType ext, TCGReg rl, } if (const_bl) { - insn = I3401_ADDSI; - if ((bl < 0) ^ sub) { - insn = I3401_SUBSI; + if (bl < 0) { bl = -bl; + insn = sub ? I3401_ADDSI : I3401_SUBSI; + } else { + insn = sub ? I3401_SUBSI : I3401_ADDSI; } + if (unlikely(al == TCG_REG_XZR)) { /* ??? We want to allow al to be zero for the benefit of negation via subtraction. However, that leaves open the -- 2.25.1 Fix a typo in the encodeing of the cmle (zero) instruction. Fixes: 14e4c1e2355 ("tcg/aarch64: Add vector operations") Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/aarch64/tcg-target.c.inc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tcg/aarch64/tcg-target.c.inc b/tcg/aarch64/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/aarch64/tcg-target.c.inc +++ b/tcg/aarch64/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ typedef enum { I3617_CMEQ0 = 0x0e209800, I3617_CMLT0 = 0x0e20a800, I3617_CMGE0 = 0x2e208800, - I3617_CMLE0 = 0x2e20a800, + I3617_CMLE0 = 0x2e209800, I3617_NOT = 0x2e205800, I3617_ABS = 0x0e20b800, I3617_NEG = 0x2e20b800, -- 2.25.1 For some vector operations, "1D" is not a valid type, and there are separate instructions for the 64-bit scalar operation. Tested-by: Stefan Weil <sw@weilnetz.de> Buglink: https://bugs.launchpad.net/qemu/+bug/1916112 Fixes: 14e4c1e2355 ("tcg/aarch64: Add vector operations") Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/aarch64/tcg-target.c.inc | 211 ++++++++++++++++++++++++++++++----- 1 file changed, 181 insertions(+), 30 deletions(-) diff --git a/tcg/aarch64/tcg-target.c.inc b/tcg/aarch64/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/aarch64/tcg-target.c.inc +++ b/tcg/aarch64/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ typedef enum { I3606_BIC = 0x2f001400, I3606_ORR = 0x0f001400, + /* AdvSIMD scalar shift by immediate */ + I3609_SSHR = 0x5f000400, + I3609_SSRA = 0x5f001400, + I3609_SHL = 0x5f005400, + I3609_USHR = 0x7f000400, + I3609_USRA = 0x7f001400, + I3609_SLI = 0x7f005400, + + /* AdvSIMD scalar three same */ + I3611_SQADD = 0x5e200c00, + I3611_SQSUB = 0x5e202c00, + I3611_CMGT = 0x5e203400, + I3611_CMGE = 0x5e203c00, + I3611_SSHL = 0x5e204400, + I3611_ADD = 0x5e208400, + I3611_CMTST = 0x5e208c00, + I3611_UQADD = 0x7e200c00, + I3611_UQSUB = 0x7e202c00, + I3611_CMHI = 0x7e203400, + I3611_CMHS = 0x7e203c00, + I3611_USHL = 0x7e204400, + I3611_SUB = 0x7e208400, + I3611_CMEQ = 0x7e208c00, + + /* AdvSIMD scalar two-reg misc */ + I3612_CMGT0 = 0x5e208800, + I3612_CMEQ0 = 0x5e209800, + I3612_CMLT0 = 0x5e20a800, + I3612_ABS = 0x5e20b800, + I3612_CMGE0 = 0x7e208800, + I3612_CMLE0 = 0x7e209800, + I3612_NEG = 0x7e20b800, + /* AdvSIMD shift by immediate */ I3614_SSHR = 0x0f000400, I3614_SSRA = 0x0f001400, @@ -XXX,XX +XXX,XX @@ static void tcg_out_insn_3606(TCGContext *s, AArch64Insn insn, bool q, | (imm8 & 0xe0) << (16 - 5) | (imm8 & 0x1f) << 5); } +static void tcg_out_insn_3609(TCGContext *s, AArch64Insn insn, + TCGReg rd, TCGReg rn, unsigned immhb) +{ + tcg_out32(s, insn | immhb << 16 | (rn & 0x1f) << 5 | (rd & 0x1f)); +} + +static void tcg_out_insn_3611(TCGContext *s, AArch64Insn insn, + unsigned size, TCGReg rd, TCGReg rn, TCGReg rm) +{ + tcg_out32(s, insn | (size << 22) | (rm & 0x1f) << 16 + | (rn & 0x1f) << 5 | (rd & 0x1f)); +} + +static void tcg_out_insn_3612(TCGContext *s, AArch64Insn insn, + unsigned size, TCGReg rd, TCGReg rn) +{ + tcg_out32(s, insn | (size << 22) | (rn & 0x1f) << 5 | (rd & 0x1f)); +} + static void tcg_out_insn_3614(TCGContext *s, AArch64Insn insn, bool q, TCGReg rd, TCGReg rn, unsigned immhb) { @@ -XXX,XX +XXX,XX @@ static void tcg_out_vec_op(TCGContext *s, TCGOpcode opc, unsigned vecl, unsigned vece, const TCGArg *args, const int *const_args) { - static const AArch64Insn cmp_insn[16] = { + static const AArch64Insn cmp_vec_insn[16] = { [TCG_COND_EQ] = I3616_CMEQ, [TCG_COND_GT] = I3616_CMGT, [TCG_COND_GE] = I3616_CMGE, [TCG_COND_GTU] = I3616_CMHI, [TCG_COND_GEU] = I3616_CMHS, }; - static const AArch64Insn cmp0_insn[16] = { + static const AArch64Insn cmp_scalar_insn[16] = { + [TCG_COND_EQ] = I3611_CMEQ, + [TCG_COND_GT] = I3611_CMGT, + [TCG_COND_GE] = I3611_CMGE, + [TCG_COND_GTU] = I3611_CMHI, + [TCG_COND_GEU] = I3611_CMHS, + }; + static const AArch64Insn cmp0_vec_insn[16] = { [TCG_COND_EQ] = I3617_CMEQ0, [TCG_COND_GT] = I3617_CMGT0, [TCG_COND_GE] = I3617_CMGE0, [TCG_COND_LT] = I3617_CMLT0, [TCG_COND_LE] = I3617_CMLE0, }; + static const AArch64Insn cmp0_scalar_insn[16] = { + [TCG_COND_EQ] = I3612_CMEQ0, + [TCG_COND_GT] = I3612_CMGT0, + [TCG_COND_GE] = I3612_CMGE0, + [TCG_COND_LT] = I3612_CMLT0, + [TCG_COND_LE] = I3612_CMLE0, + }; TCGType type = vecl + TCG_TYPE_V64; unsigned is_q = vecl; + bool is_scalar = !is_q && vece == MO_64; TCGArg a0, a1, a2, a3; int cmode, imm8; @@ -XXX,XX +XXX,XX @@ static void tcg_out_vec_op(TCGContext *s, TCGOpcode opc, tcg_out_dupm_vec(s, type, vece, a0, a1, a2); break; case INDEX_op_add_vec: - tcg_out_insn(s, 3616, ADD, is_q, vece, a0, a1, a2); + if (is_scalar) { + tcg_out_insn(s, 3611, ADD, vece, a0, a1, a2); + } else { + tcg_out_insn(s, 3616, ADD, is_q, vece, a0, a1, a2); + } break; case INDEX_op_sub_vec: - tcg_out_insn(s, 3616, SUB, is_q, vece, a0, a1, a2); + if (is_scalar) { + tcg_out_insn(s, 3611, SUB, vece, a0, a1, a2); + } else { + tcg_out_insn(s, 3616, SUB, is_q, vece, a0, a1, a2); + } break; case INDEX_op_mul_vec: tcg_out_insn(s, 3616, MUL, is_q, vece, a0, a1, a2); break; case INDEX_op_neg_vec: - tcg_out_insn(s, 3617, NEG, is_q, vece, a0, a1); + if (is_scalar) { + tcg_out_insn(s, 3612, NEG, vece, a0, a1); + } else { + tcg_out_insn(s, 3617, NEG, is_q, vece, a0, a1); + } break; case INDEX_op_abs_vec: - tcg_out_insn(s, 3617, ABS, is_q, vece, a0, a1); + if (is_scalar) { + tcg_out_insn(s, 3612, ABS, vece, a0, a1); + } else { + tcg_out_insn(s, 3617, ABS, is_q, vece, a0, a1); + } break; case INDEX_op_and_vec: if (const_args[2]) { @@ -XXX,XX +XXX,XX @@ static void tcg_out_vec_op(TCGContext *s, TCGOpcode opc, tcg_out_insn(s, 3616, EOR, is_q, 0, a0, a1, a2); break; case INDEX_op_ssadd_vec: - tcg_out_insn(s, 3616, SQADD, is_q, vece, a0, a1, a2); + if (is_scalar) { + tcg_out_insn(s, 3611, SQADD, vece, a0, a1, a2); + } else { + tcg_out_insn(s, 3616, SQADD, is_q, vece, a0, a1, a2); + } break; case INDEX_op_sssub_vec: - tcg_out_insn(s, 3616, SQSUB, is_q, vece, a0, a1, a2); + if (is_scalar) { + tcg_out_insn(s, 3611, SQSUB, vece, a0, a1, a2); + } else { + tcg_out_insn(s, 3616, SQSUB, is_q, vece, a0, a1, a2); + } break; case INDEX_op_usadd_vec: - tcg_out_insn(s, 3616, UQADD, is_q, vece, a0, a1, a2); + if (is_scalar) { + tcg_out_insn(s, 3611, UQADD, vece, a0, a1, a2); + } else { + tcg_out_insn(s, 3616, UQADD, is_q, vece, a0, a1, a2); + } break; case INDEX_op_ussub_vec: - tcg_out_insn(s, 3616, UQSUB, is_q, vece, a0, a1, a2); + if (is_scalar) { + tcg_out_insn(s, 3611, UQSUB, vece, a0, a1, a2); + } else { + tcg_out_insn(s, 3616, UQSUB, is_q, vece, a0, a1, a2); + } break; case INDEX_op_smax_vec: tcg_out_insn(s, 3616, SMAX, is_q, vece, a0, a1, a2); @@ -XXX,XX +XXX,XX @@ static void tcg_out_vec_op(TCGContext *s, TCGOpcode opc, tcg_out_insn(s, 3617, NOT, is_q, 0, a0, a1); break; case INDEX_op_shli_vec: - tcg_out_insn(s, 3614, SHL, is_q, a0, a1, a2 + (8 << vece)); + if (is_scalar) { + tcg_out_insn(s, 3609, SHL, a0, a1, a2 + (8 << vece)); + } else { + tcg_out_insn(s, 3614, SHL, is_q, a0, a1, a2 + (8 << vece)); + } break; case INDEX_op_shri_vec: - tcg_out_insn(s, 3614, USHR, is_q, a0, a1, (16 << vece) - a2); + if (is_scalar) { + tcg_out_insn(s, 3609, USHR, a0, a1, (16 << vece) - a2); + } else { + tcg_out_insn(s, 3614, USHR, is_q, a0, a1, (16 << vece) - a2); + } break; case INDEX_op_sari_vec: - tcg_out_insn(s, 3614, SSHR, is_q, a0, a1, (16 << vece) - a2); + if (is_scalar) { + tcg_out_insn(s, 3609, SSHR, a0, a1, (16 << vece) - a2); + } else { + tcg_out_insn(s, 3614, SSHR, is_q, a0, a1, (16 << vece) - a2); + } break; case INDEX_op_aa64_sli_vec: - tcg_out_insn(s, 3614, SLI, is_q, a0, a2, args[3] + (8 << vece)); + if (is_scalar) { + tcg_out_insn(s, 3609, SLI, a0, a2, args[3] + (8 << vece)); + } else { + tcg_out_insn(s, 3614, SLI, is_q, a0, a2, args[3] + (8 << vece)); + } break; case INDEX_op_shlv_vec: - tcg_out_insn(s, 3616, USHL, is_q, vece, a0, a1, a2); + if (is_scalar) { + tcg_out_insn(s, 3611, USHL, vece, a0, a1, a2); + } else { + tcg_out_insn(s, 3616, USHL, is_q, vece, a0, a1, a2); + } break; case INDEX_op_aa64_sshl_vec: - tcg_out_insn(s, 3616, SSHL, is_q, vece, a0, a1, a2); + if (is_scalar) { + tcg_out_insn(s, 3611, SSHL, vece, a0, a1, a2); + } else { + tcg_out_insn(s, 3616, SSHL, is_q, vece, a0, a1, a2); + } break; case INDEX_op_cmp_vec: { @@ -XXX,XX +XXX,XX @@ static void tcg_out_vec_op(TCGContext *s, TCGOpcode opc, if (cond == TCG_COND_NE) { if (const_args[2]) { - tcg_out_insn(s, 3616, CMTST, is_q, vece, a0, a1, a1); + if (is_scalar) { + tcg_out_insn(s, 3611, CMTST, vece, a0, a1, a1); + } else { + tcg_out_insn(s, 3616, CMTST, is_q, vece, a0, a1, a1); + } } else { - tcg_out_insn(s, 3616, CMEQ, is_q, vece, a0, a1, a2); + if (is_scalar) { + tcg_out_insn(s, 3611, CMEQ, vece, a0, a1, a2); + } else { + tcg_out_insn(s, 3616, CMEQ, is_q, vece, a0, a1, a2); + } tcg_out_insn(s, 3617, NOT, is_q, 0, a0, a0); } } else { if (const_args[2]) { - insn = cmp0_insn[cond]; - if (insn) { - tcg_out_insn_3617(s, insn, is_q, vece, a0, a1); - break; + if (is_scalar) { + insn = cmp0_scalar_insn[cond]; + if (insn) { + tcg_out_insn_3612(s, insn, vece, a0, a1); + break; + } + } else { + insn = cmp0_vec_insn[cond]; + if (insn) { + tcg_out_insn_3617(s, insn, is_q, vece, a0, a1); + break; + } } tcg_out_dupi_vec(s, type, MO_8, TCG_VEC_TMP, 0); a2 = TCG_VEC_TMP; } - insn = cmp_insn[cond]; - if (insn == 0) { - TCGArg t; - t = a1, a1 = a2, a2 = t; - cond = tcg_swap_cond(cond); - insn = cmp_insn[cond]; - tcg_debug_assert(insn != 0); + if (is_scalar) { + insn = cmp_scalar_insn[cond]; + if (insn == 0) { + TCGArg t; + t = a1, a1 = a2, a2 = t; + cond = tcg_swap_cond(cond); + insn = cmp_scalar_insn[cond]; + tcg_debug_assert(insn != 0); + } + tcg_out_insn_3611(s, insn, vece, a0, a1, a2); + } else { + insn = cmp_vec_insn[cond]; + if (insn == 0) { + TCGArg t; + t = a1, a1 = a2, a2 = t; + cond = tcg_swap_cond(cond); + insn = cmp_vec_insn[cond]; + tcg_debug_assert(insn != 0); + } + tcg_out_insn_3616(s, insn, is_q, vece, a0, a1, a2); } - tcg_out_insn_3616(s, insn, is_q, vece, a0, a1, a2); } } break; -- 2.25.1 Use the provided cpu_ldst.h interfaces. This fixes the build vs the unconverted uses of g2h(), adds missed memory trace events, and correctly recognizes when a SIGSEGV belongs to the guest via set_helper_retaddr(). Fixes: 3e8f1628e864 Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 73 +++++++++++++++++++++---------------------------------- 1 file changed, 28 insertions(+), 45 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ static bool tci_compare64(uint64_t u0, uint64_t u1, TCGCond condition) return result; } -#ifdef CONFIG_SOFTMMU -# define qemu_ld_ub \ - helper_ret_ldub_mmu(env, taddr, oi, (uintptr_t)tb_ptr) -# define qemu_ld_leuw \ - helper_le_lduw_mmu(env, taddr, oi, (uintptr_t)tb_ptr) -# define qemu_ld_leul \ - helper_le_ldul_mmu(env, taddr, oi, (uintptr_t)tb_ptr) -# define qemu_ld_leq \ - helper_le_ldq_mmu(env, taddr, oi, (uintptr_t)tb_ptr) -# define qemu_ld_beuw \ - helper_be_lduw_mmu(env, taddr, oi, (uintptr_t)tb_ptr) -# define qemu_ld_beul \ - helper_be_ldul_mmu(env, taddr, oi, (uintptr_t)tb_ptr) -# define qemu_ld_beq \ - helper_be_ldq_mmu(env, taddr, oi, (uintptr_t)tb_ptr) -# define qemu_st_b(X) \ - helper_ret_stb_mmu(env, taddr, X, oi, (uintptr_t)tb_ptr) -# define qemu_st_lew(X) \ - helper_le_stw_mmu(env, taddr, X, oi, (uintptr_t)tb_ptr) -# define qemu_st_lel(X) \ - helper_le_stl_mmu(env, taddr, X, oi, (uintptr_t)tb_ptr) -# define qemu_st_leq(X) \ - helper_le_stq_mmu(env, taddr, X, oi, (uintptr_t)tb_ptr) -# define qemu_st_bew(X) \ - helper_be_stw_mmu(env, taddr, X, oi, (uintptr_t)tb_ptr) -# define qemu_st_bel(X) \ - helper_be_stl_mmu(env, taddr, X, oi, (uintptr_t)tb_ptr) -# define qemu_st_beq(X) \ - helper_be_stq_mmu(env, taddr, X, oi, (uintptr_t)tb_ptr) -#else -# define qemu_ld_ub ldub_p(g2h(taddr)) -# define qemu_ld_leuw lduw_le_p(g2h(taddr)) -# define qemu_ld_leul (uint32_t)ldl_le_p(g2h(taddr)) -# define qemu_ld_leq ldq_le_p(g2h(taddr)) -# define qemu_ld_beuw lduw_be_p(g2h(taddr)) -# define qemu_ld_beul (uint32_t)ldl_be_p(g2h(taddr)) -# define qemu_ld_beq ldq_be_p(g2h(taddr)) -# define qemu_st_b(X) stb_p(g2h(taddr), X) -# define qemu_st_lew(X) stw_le_p(g2h(taddr), X) -# define qemu_st_lel(X) stl_le_p(g2h(taddr), X) -# define qemu_st_leq(X) stq_le_p(g2h(taddr), X) -# define qemu_st_bew(X) stw_be_p(g2h(taddr), X) -# define qemu_st_bel(X) stl_be_p(g2h(taddr), X) -# define qemu_st_beq(X) stq_be_p(g2h(taddr), X) -#endif +#define qemu_ld_ub \ + cpu_ldub_mmuidx_ra(env, taddr, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_ld_leuw \ + cpu_lduw_le_mmuidx_ra(env, taddr, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_ld_leul \ + cpu_ldl_le_mmuidx_ra(env, taddr, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_ld_leq \ + cpu_ldq_le_mmuidx_ra(env, taddr, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_ld_beuw \ + cpu_lduw_be_mmuidx_ra(env, taddr, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_ld_beul \ + cpu_ldl_be_mmuidx_ra(env, taddr, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_ld_beq \ + cpu_ldq_be_mmuidx_ra(env, taddr, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_st_b(X) \ + cpu_stb_mmuidx_ra(env, taddr, X, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_st_lew(X) \ + cpu_stw_le_mmuidx_ra(env, taddr, X, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_st_lel(X) \ + cpu_stl_le_mmuidx_ra(env, taddr, X, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_st_leq(X) \ + cpu_stq_le_mmuidx_ra(env, taddr, X, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_st_bew(X) \ + cpu_stw_be_mmuidx_ra(env, taddr, X, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_st_bel(X) \ + cpu_stl_be_mmuidx_ra(env, taddr, X, get_mmuidx(oi), (uintptr_t)tb_ptr) +#define qemu_st_beq(X) \ + cpu_stq_be_mmuidx_ra(env, taddr, X, get_mmuidx(oi), (uintptr_t)tb_ptr) #if TCG_TARGET_REG_BITS == 64 # define CASE_32_64(x) \ -- 2.25.1 Allow other places in tcg to restart with a smaller tb. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tcg.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ static void set_jmp_reset_offset(TCGContext *s, int which) s->tb_jmp_reset_offset[which] = tcg_current_code_size(s); } +/* Signal overflow, starting over with fewer guest insns. */ +static void QEMU_NORETURN tcg_raise_tb_overflow(TCGContext *s) +{ + siglongjmp(s->jmp_trans, -2); +} + #define C_PFX1(P, A) P##A #define C_PFX2(P, A, B) P##A##_##B #define C_PFX3(P, A, B, C) P##A##_##B##_##C @@ -XXX,XX +XXX,XX @@ static TCGTemp *tcg_temp_alloc(TCGContext *s) int n = s->nb_temps++; if (n >= TCG_MAX_TEMPS) { - /* Signal overflow, starting over with fewer guest insns. */ - siglongjmp(s->jmp_trans, -2); + tcg_raise_tb_overflow(s); } return memset(&s->temps[n], 0, sizeof(TCGTemp)); } -- 2.25.1 The use in tcg_tb_lookup is given a random pc that comes from the pc of a signal handler. Do not assert that the pointer is already within the code gen buffer at all, much less the writable mirror of it. Fixes: db0c51a3803 Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tcg.c | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ static void tcg_region_trees_init(void) } } -static struct tcg_region_tree *tc_ptr_to_region_tree(const void *cp) +static struct tcg_region_tree *tc_ptr_to_region_tree(const void *p) { - void *p = tcg_splitwx_to_rw(cp); size_t region_idx; + /* + * Like tcg_splitwx_to_rw, with no assert. The pc may come from + * a signal handler over which the caller has no control. + */ + if (!in_code_gen_buffer(p)) { + p -= tcg_splitwx_diff; + if (!in_code_gen_buffer(p)) { + return NULL; + } + } + if (p < region.start_aligned) { region_idx = 0; } else { @@ -XXX,XX +XXX,XX @@ void tcg_tb_insert(TranslationBlock *tb) { struct tcg_region_tree *rt = tc_ptr_to_region_tree(tb->tc.ptr); + g_assert(rt != NULL); qemu_mutex_lock(&rt->lock); g_tree_insert(rt->tree, &tb->tc, tb); qemu_mutex_unlock(&rt->lock); @@ -XXX,XX +XXX,XX @@ void tcg_tb_remove(TranslationBlock *tb) { struct tcg_region_tree *rt = tc_ptr_to_region_tree(tb->tc.ptr); + g_assert(rt != NULL); qemu_mutex_lock(&rt->lock); g_tree_remove(rt->tree, &tb->tc); qemu_mutex_unlock(&rt->lock); @@ -XXX,XX +XXX,XX @@ TranslationBlock *tcg_tb_lookup(uintptr_t tc_ptr) TranslationBlock *tb; struct tb_tc s = { .ptr = (void *)tc_ptr }; + if (rt == NULL) { + return NULL; + } + qemu_mutex_lock(&rt->lock); tb = g_tree_lookup(rt->tree, &s); qemu_mutex_unlock(&rt->lock); -- 2.25.1 Use CASE_32_64 and CASE_64 to reduce ifdefs and merge cases that are identical between 32-bit and 64-bit hosts. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210217202036.1724901-5-richard.henderson@linaro.org> [PMD: Split patch as 1/5] Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210218232840.1760806-2-f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci/tcg-target.c.inc | 85 +++++++++++++++++----------------------- 1 file changed, 37 insertions(+), 48 deletions(-) diff --git a/tcg/tci/tcg-target.c.inc b/tcg/tci/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci/tcg-target.c.inc +++ b/tcg/tci/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static inline void tcg_out_call(TCGContext *s, const tcg_insn_unit *arg) old_code_ptr[1] = s->code_ptr - old_code_ptr; } +#if TCG_TARGET_REG_BITS == 64 +# define CASE_32_64(x) \ + case glue(glue(INDEX_op_, x), _i64): \ + case glue(glue(INDEX_op_, x), _i32): +# define CASE_64(x) \ + case glue(glue(INDEX_op_, x), _i64): +#else +# define CASE_32_64(x) \ + case glue(glue(INDEX_op_, x), _i32): +# define CASE_64(x) +#endif + static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, const int *const_args) { @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, case INDEX_op_exit_tb: tcg_out64(s, args[0]); break; + case INDEX_op_goto_tb: if (s->tb_jmp_insn_offset) { /* Direct jump method. */ @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, tcg_debug_assert(args[2] == (int32_t)args[2]); tcg_out32(s, args[2]); break; - case INDEX_op_add_i32: - case INDEX_op_sub_i32: - case INDEX_op_mul_i32: - case INDEX_op_and_i32: - case INDEX_op_andc_i32: /* Optional (TCG_TARGET_HAS_andc_i32). */ - case INDEX_op_eqv_i32: /* Optional (TCG_TARGET_HAS_eqv_i32). */ - case INDEX_op_nand_i32: /* Optional (TCG_TARGET_HAS_nand_i32). */ - case INDEX_op_nor_i32: /* Optional (TCG_TARGET_HAS_nor_i32). */ - case INDEX_op_or_i32: - case INDEX_op_orc_i32: /* Optional (TCG_TARGET_HAS_orc_i32). */ - case INDEX_op_xor_i32: - case INDEX_op_shl_i32: - case INDEX_op_shr_i32: - case INDEX_op_sar_i32: - case INDEX_op_rotl_i32: /* Optional (TCG_TARGET_HAS_rot_i32). */ - case INDEX_op_rotr_i32: /* Optional (TCG_TARGET_HAS_rot_i32). */ + + CASE_32_64(add) + CASE_32_64(sub) + CASE_32_64(mul) + CASE_32_64(and) + CASE_32_64(or) + CASE_32_64(xor) + CASE_32_64(andc) /* Optional (TCG_TARGET_HAS_andc_*). */ + CASE_32_64(orc) /* Optional (TCG_TARGET_HAS_orc_*). */ + CASE_32_64(eqv) /* Optional (TCG_TARGET_HAS_eqv_*). */ + CASE_32_64(nand) /* Optional (TCG_TARGET_HAS_nand_*). */ + CASE_32_64(nor) /* Optional (TCG_TARGET_HAS_nor_*). */ + CASE_32_64(shl) + CASE_32_64(shr) + CASE_32_64(sar) + CASE_32_64(rotl) /* Optional (TCG_TARGET_HAS_rot_*). */ + CASE_32_64(rotr) /* Optional (TCG_TARGET_HAS_rot_*). */ + CASE_32_64(div) /* Optional (TCG_TARGET_HAS_div_*). */ + CASE_32_64(divu) /* Optional (TCG_TARGET_HAS_div_*). */ + CASE_32_64(rem) /* Optional (TCG_TARGET_HAS_div_*). */ + CASE_32_64(remu) /* Optional (TCG_TARGET_HAS_div_*). */ tcg_out_r(s, args[0]); tcg_out_r(s, args[1]); tcg_out_r(s, args[2]); @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, break; #if TCG_TARGET_REG_BITS == 64 - case INDEX_op_add_i64: - case INDEX_op_sub_i64: - case INDEX_op_mul_i64: - case INDEX_op_and_i64: - case INDEX_op_andc_i64: /* Optional (TCG_TARGET_HAS_andc_i64). */ - case INDEX_op_eqv_i64: /* Optional (TCG_TARGET_HAS_eqv_i64). */ - case INDEX_op_nand_i64: /* Optional (TCG_TARGET_HAS_nand_i64). */ - case INDEX_op_nor_i64: /* Optional (TCG_TARGET_HAS_nor_i64). */ - case INDEX_op_or_i64: - case INDEX_op_orc_i64: /* Optional (TCG_TARGET_HAS_orc_i64). */ - case INDEX_op_xor_i64: - case INDEX_op_shl_i64: - case INDEX_op_shr_i64: - case INDEX_op_sar_i64: - case INDEX_op_rotl_i64: /* Optional (TCG_TARGET_HAS_rot_i64). */ - case INDEX_op_rotr_i64: /* Optional (TCG_TARGET_HAS_rot_i64). */ - case INDEX_op_div_i64: /* Optional (TCG_TARGET_HAS_div_i64). */ - case INDEX_op_divu_i64: /* Optional (TCG_TARGET_HAS_div_i64). */ - case INDEX_op_rem_i64: /* Optional (TCG_TARGET_HAS_div_i64). */ - case INDEX_op_remu_i64: /* Optional (TCG_TARGET_HAS_div_i64). */ - tcg_out_r(s, args[0]); - tcg_out_r(s, args[1]); - tcg_out_r(s, args[2]); - break; case INDEX_op_deposit_i64: /* Optional (TCG_TARGET_HAS_deposit_i64). */ tcg_out_r(s, args[0]); tcg_out_r(s, args[1]); @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, tcg_out_r(s, args[0]); tcg_out_r(s, args[1]); break; - case INDEX_op_div_i32: /* Optional (TCG_TARGET_HAS_div_i32). */ - case INDEX_op_divu_i32: /* Optional (TCG_TARGET_HAS_div_i32). */ - case INDEX_op_rem_i32: /* Optional (TCG_TARGET_HAS_div_i32). */ - case INDEX_op_remu_i32: /* Optional (TCG_TARGET_HAS_div_i32). */ - tcg_out_r(s, args[0]); - tcg_out_r(s, args[1]); - tcg_out_r(s, args[2]); - break; + #if TCG_TARGET_REG_BITS == 32 case INDEX_op_add2_i32: case INDEX_op_sub2_i32: @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, } tcg_out_i(s, *args++); break; + case INDEX_op_mb: break; + case INDEX_op_mov_i32: /* Always emitted via tcg_out_mov. */ case INDEX_op_mov_i64: case INDEX_op_call: /* Always emitted via tcg_out_call. */ -- 2.25.1 Use CASE_32_64 and CASE_64 to reduce ifdefs and merge cases that are identical between 32-bit and 64-bit hosts. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210217202036.1724901-5-richard.henderson@linaro.org> [PMD: Split patch as 2/5] Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210218232840.1760806-3-f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci/tcg-target.c.inc | 35 ++++++++++++++--------------------- 1 file changed, 14 insertions(+), 21 deletions(-) diff --git a/tcg/tci/tcg-target.c.inc b/tcg/tci/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci/tcg-target.c.inc +++ b/tcg/tci/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, tcg_out8(s, args[2]); /* condition */ tci_out_label(s, arg_label(args[3])); break; - case INDEX_op_bswap16_i64: /* Optional (TCG_TARGET_HAS_bswap16_i64). */ - case INDEX_op_bswap32_i64: /* Optional (TCG_TARGET_HAS_bswap32_i64). */ - case INDEX_op_bswap64_i64: /* Optional (TCG_TARGET_HAS_bswap64_i64). */ - case INDEX_op_not_i64: /* Optional (TCG_TARGET_HAS_not_i64). */ - case INDEX_op_neg_i64: /* Optional (TCG_TARGET_HAS_neg_i64). */ - case INDEX_op_ext8s_i64: /* Optional (TCG_TARGET_HAS_ext8s_i64). */ - case INDEX_op_ext8u_i64: /* Optional (TCG_TARGET_HAS_ext8u_i64). */ - case INDEX_op_ext16s_i64: /* Optional (TCG_TARGET_HAS_ext16s_i64). */ - case INDEX_op_ext16u_i64: /* Optional (TCG_TARGET_HAS_ext16u_i64). */ - case INDEX_op_ext32s_i64: /* Optional (TCG_TARGET_HAS_ext32s_i64). */ - case INDEX_op_ext32u_i64: /* Optional (TCG_TARGET_HAS_ext32u_i64). */ - case INDEX_op_ext_i32_i64: - case INDEX_op_extu_i32_i64: #endif /* TCG_TARGET_REG_BITS == 64 */ - case INDEX_op_neg_i32: /* Optional (TCG_TARGET_HAS_neg_i32). */ - case INDEX_op_not_i32: /* Optional (TCG_TARGET_HAS_not_i32). */ - case INDEX_op_ext8s_i32: /* Optional (TCG_TARGET_HAS_ext8s_i32). */ - case INDEX_op_ext16s_i32: /* Optional (TCG_TARGET_HAS_ext16s_i32). */ - case INDEX_op_ext8u_i32: /* Optional (TCG_TARGET_HAS_ext8u_i32). */ - case INDEX_op_ext16u_i32: /* Optional (TCG_TARGET_HAS_ext16u_i32). */ - case INDEX_op_bswap16_i32: /* Optional (TCG_TARGET_HAS_bswap16_i32). */ - case INDEX_op_bswap32_i32: /* Optional (TCG_TARGET_HAS_bswap32_i32). */ + + CASE_32_64(neg) /* Optional (TCG_TARGET_HAS_neg_*). */ + CASE_32_64(not) /* Optional (TCG_TARGET_HAS_not_*). */ + CASE_32_64(ext8s) /* Optional (TCG_TARGET_HAS_ext8s_*). */ + CASE_32_64(ext8u) /* Optional (TCG_TARGET_HAS_ext8u_*). */ + CASE_32_64(ext16s) /* Optional (TCG_TARGET_HAS_ext16s_*). */ + CASE_32_64(ext16u) /* Optional (TCG_TARGET_HAS_ext16u_*). */ + CASE_64(ext32s) /* Optional (TCG_TARGET_HAS_ext32s_i64). */ + CASE_64(ext32u) /* Optional (TCG_TARGET_HAS_ext32u_i64). */ + CASE_64(ext_i32) + CASE_64(extu_i32) + CASE_32_64(bswap16) /* Optional (TCG_TARGET_HAS_bswap16_*). */ + CASE_32_64(bswap32) /* Optional (TCG_TARGET_HAS_bswap32_*). */ + CASE_64(bswap64) /* Optional (TCG_TARGET_HAS_bswap64_i64). */ tcg_out_r(s, args[0]); tcg_out_r(s, args[1]); break; -- 2.25.1 Use CASE_32_64 and CASE_64 to reduce ifdefs and merge cases that are identical between 32-bit and 64-bit hosts. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210217202036.1724901-5-richard.henderson@linaro.org> [PMD: Split patch as 3/5] Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210218232840.1760806-4-f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci/tcg-target.c.inc | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) diff --git a/tcg/tci/tcg-target.c.inc b/tcg/tci/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci/tcg-target.c.inc +++ b/tcg/tci/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, tcg_out_r(s, args[1]); tcg_out_r(s, args[2]); break; - case INDEX_op_deposit_i32: /* Optional (TCG_TARGET_HAS_deposit_i32). */ + + CASE_32_64(deposit) /* Optional (TCG_TARGET_HAS_deposit_*). */ tcg_out_r(s, args[0]); tcg_out_r(s, args[1]); tcg_out_r(s, args[2]); @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, break; #if TCG_TARGET_REG_BITS == 64 - case INDEX_op_deposit_i64: /* Optional (TCG_TARGET_HAS_deposit_i64). */ - tcg_out_r(s, args[0]); - tcg_out_r(s, args[1]); - tcg_out_r(s, args[2]); - tcg_debug_assert(args[3] <= UINT8_MAX); - tcg_out8(s, args[3]); - tcg_debug_assert(args[4] <= UINT8_MAX); - tcg_out8(s, args[4]); - break; case INDEX_op_brcond_i64: tcg_out_r(s, args[0]); tcg_out_r(s, args[1]); -- 2.25.1 Use CASE_32_64 and CASE_64 to reduce ifdefs and merge cases that are identical between 32-bit and 64-bit hosts. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210217202036.1724901-5-richard.henderson@linaro.org> [PMD: Split patch as 4/5] Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210218232840.1760806-5-f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci/tcg-target.c.inc | 23 ++++++----------------- 1 file changed, 6 insertions(+), 17 deletions(-) diff --git a/tcg/tci/tcg-target.c.inc b/tcg/tci/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci/tcg-target.c.inc +++ b/tcg/tci/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, } set_jmp_reset_offset(s, args[0]); break; + case INDEX_op_br: tci_out_label(s, arg_label(args[0])); break; - case INDEX_op_setcond_i32: + + CASE_32_64(setcond) tcg_out_r(s, args[0]); tcg_out_r(s, args[1]); tcg_out_r(s, args[2]); tcg_out8(s, args[3]); /* condition */ break; + #if TCG_TARGET_REG_BITS == 32 case INDEX_op_setcond2_i32: /* setcond2_i32 cond, t0, t1_low, t1_high, t2_low, t2_high */ @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, tcg_out_r(s, args[4]); tcg_out8(s, args[5]); /* condition */ break; -#elif TCG_TARGET_REG_BITS == 64 - case INDEX_op_setcond_i64: - tcg_out_r(s, args[0]); - tcg_out_r(s, args[1]); - tcg_out_r(s, args[2]); - tcg_out8(s, args[3]); /* condition */ - break; #endif case INDEX_op_ld8u_i32: case INDEX_op_ld8s_i32: @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, tcg_out8(s, args[4]); break; -#if TCG_TARGET_REG_BITS == 64 - case INDEX_op_brcond_i64: + CASE_32_64(brcond) tcg_out_r(s, args[0]); tcg_out_r(s, args[1]); tcg_out8(s, args[2]); /* condition */ tci_out_label(s, arg_label(args[3])); break; -#endif /* TCG_TARGET_REG_BITS == 64 */ CASE_32_64(neg) /* Optional (TCG_TARGET_HAS_neg_*). */ CASE_32_64(not) /* Optional (TCG_TARGET_HAS_not_*). */ @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, tcg_out_r(s, args[3]); break; #endif - case INDEX_op_brcond_i32: - tcg_out_r(s, args[0]); - tcg_out_r(s, args[1]); - tcg_out8(s, args[2]); /* condition */ - tci_out_label(s, arg_label(args[3])); - break; + case INDEX_op_qemu_ld_i32: tcg_out_r(s, *args++); tcg_out_r(s, *args++); -- 2.25.1 Use CASE_32_64 and CASE_64 to reduce ifdefs and merge cases that are identical between 32-bit and 64-bit hosts. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210217202036.1724901-5-richard.henderson@linaro.org> [PMD: Split patch as 5/5] Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-Id: <20210218232840.1760806-6-f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci/tcg-target.c.inc | 49 ++++++++++++---------------------------- 1 file changed, 14 insertions(+), 35 deletions(-) diff --git a/tcg/tci/tcg-target.c.inc b/tcg/tci/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci/tcg-target.c.inc +++ b/tcg/tci/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, tcg_out8(s, args[5]); /* condition */ break; #endif - case INDEX_op_ld8u_i32: - case INDEX_op_ld8s_i32: - case INDEX_op_ld16u_i32: - case INDEX_op_ld16s_i32: + + CASE_32_64(ld8u) + CASE_32_64(ld8s) + CASE_32_64(ld16u) + CASE_32_64(ld16s) case INDEX_op_ld_i32: - case INDEX_op_st8_i32: - case INDEX_op_st16_i32: + CASE_64(ld32u) + CASE_64(ld32s) + CASE_64(ld) + CASE_32_64(st8) + CASE_32_64(st16) case INDEX_op_st_i32: - case INDEX_op_ld8u_i64: - case INDEX_op_ld8s_i64: - case INDEX_op_ld16u_i64: - case INDEX_op_ld16s_i64: - case INDEX_op_ld32u_i64: - case INDEX_op_ld32s_i64: - case INDEX_op_ld_i64: - case INDEX_op_st8_i64: - case INDEX_op_st16_i64: - case INDEX_op_st32_i64: - case INDEX_op_st_i64: + CASE_64(st32) + CASE_64(st) stack_bounds_check(args[1], args[2]); tcg_out_r(s, args[0]); tcg_out_r(s, args[1]); @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, #endif case INDEX_op_qemu_ld_i32: - tcg_out_r(s, *args++); - tcg_out_r(s, *args++); - if (TARGET_LONG_BITS > TCG_TARGET_REG_BITS) { - tcg_out_r(s, *args++); - } - tcg_out_i(s, *args++); - break; - case INDEX_op_qemu_ld_i64: - tcg_out_r(s, *args++); - if (TCG_TARGET_REG_BITS == 32) { - tcg_out_r(s, *args++); - } - tcg_out_r(s, *args++); - if (TARGET_LONG_BITS > TCG_TARGET_REG_BITS) { - tcg_out_r(s, *args++); - } - tcg_out_i(s, *args++); - break; case INDEX_op_qemu_st_i32: tcg_out_r(s, *args++); tcg_out_r(s, *args++); @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, } tcg_out_i(s, *args++); break; + + case INDEX_op_qemu_ld_i64: case INDEX_op_qemu_st_i64: tcg_out_r(s, *args++); if (TCG_TARGET_REG_BITS == 32) { -- 2.25.1 Use explicit casts for ext8u opcodes, and allow truncation to happen with the store for st8 opcodes. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 23 +++++------------------ 1 file changed, 5 insertions(+), 18 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ static int32_t tci_read_reg32s(const tcg_target_ulong *regs, TCGReg index) } #endif -static uint8_t tci_read_reg8(const tcg_target_ulong *regs, TCGReg index) -{ - return (uint8_t)tci_read_reg(regs, index); -} - static uint16_t tci_read_reg16(const tcg_target_ulong *regs, TCGReg index) { return (uint16_t)tci_read_reg(regs, index); @@ -XXX,XX +XXX,XX @@ tci_read_r(const tcg_target_ulong *regs, const uint8_t **tb_ptr) return value; } -/* Read indexed register (8 bit) from bytecode. */ -static uint8_t tci_read_r8(const tcg_target_ulong *regs, const uint8_t **tb_ptr) -{ - uint8_t value = tci_read_reg8(regs, **tb_ptr); - *tb_ptr += 1; - return value; -} - #if TCG_TARGET_HAS_ext8s_i32 || TCG_TARGET_HAS_ext8s_i64 /* Read indexed register (8 bit signed) from bytecode. */ static int8_t tci_read_r8s(const tcg_target_ulong *regs, const uint8_t **tb_ptr) @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, tci_write_reg(regs, t0, *(uint32_t *)(t1 + t2)); break; CASE_32_64(st8) - t0 = tci_read_r8(regs, &tb_ptr); + t0 = tci_read_r(regs, &tb_ptr); t1 = tci_read_r(regs, &tb_ptr); t2 = tci_read_s32(&tb_ptr); *(uint8_t *)(t1 + t2) = t0; @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_ext8u_i32 case INDEX_op_ext8u_i32: t0 = *tb_ptr++; - t1 = tci_read_r8(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (uint8_t)t1); break; #endif #if TCG_TARGET_HAS_ext16u_i32 @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_ext8u_i64 case INDEX_op_ext8u_i64: t0 = *tb_ptr++; - t1 = tci_read_r8(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (uint8_t)t1); break; #endif #if TCG_TARGET_HAS_ext8s_i64 -- 2.25.1 Use explicit casts for ext8s opcodes. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 25 ++++--------------------- 1 file changed, 4 insertions(+), 21 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ static tcg_target_ulong tci_read_reg(const tcg_target_ulong *regs, TCGReg index) return regs[index]; } -#if TCG_TARGET_HAS_ext8s_i32 || TCG_TARGET_HAS_ext8s_i64 -static int8_t tci_read_reg8s(const tcg_target_ulong *regs, TCGReg index) -{ - return (int8_t)tci_read_reg(regs, index); -} -#endif - #if TCG_TARGET_HAS_ext16s_i32 || TCG_TARGET_HAS_ext16s_i64 static int16_t tci_read_reg16s(const tcg_target_ulong *regs, TCGReg index) { @@ -XXX,XX +XXX,XX @@ tci_read_r(const tcg_target_ulong *regs, const uint8_t **tb_ptr) return value; } -#if TCG_TARGET_HAS_ext8s_i32 || TCG_TARGET_HAS_ext8s_i64 -/* Read indexed register (8 bit signed) from bytecode. */ -static int8_t tci_read_r8s(const tcg_target_ulong *regs, const uint8_t **tb_ptr) -{ - int8_t value = tci_read_reg8s(regs, **tb_ptr); - *tb_ptr += 1; - return value; -} -#endif - /* Read indexed register (16 bit) from bytecode. */ static uint16_t tci_read_r16(const tcg_target_ulong *regs, const uint8_t **tb_ptr) @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_ext8s_i32 case INDEX_op_ext8s_i32: t0 = *tb_ptr++; - t1 = tci_read_r8s(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (int8_t)t1); break; #endif #if TCG_TARGET_HAS_ext16s_i32 @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_ext8s_i64 case INDEX_op_ext8s_i64: t0 = *tb_ptr++; - t1 = tci_read_r8s(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (int8_t)t1); break; #endif #if TCG_TARGET_HAS_ext16s_i64 -- 2.25.1 Use explicit casts for ext16u opcodes, and allow truncation to happen with the store for st16 opcodes, and with the call for bswap16 opcodes. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 28 +++++++--------------------- 1 file changed, 7 insertions(+), 21 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ static int32_t tci_read_reg32s(const tcg_target_ulong *regs, TCGReg index) } #endif -static uint16_t tci_read_reg16(const tcg_target_ulong *regs, TCGReg index) -{ - return (uint16_t)tci_read_reg(regs, index); -} - static uint32_t tci_read_reg32(const tcg_target_ulong *regs, TCGReg index) { return (uint32_t)tci_read_reg(regs, index); @@ -XXX,XX +XXX,XX @@ tci_read_r(const tcg_target_ulong *regs, const uint8_t **tb_ptr) return value; } -/* Read indexed register (16 bit) from bytecode. */ -static uint16_t tci_read_r16(const tcg_target_ulong *regs, - const uint8_t **tb_ptr) -{ - uint16_t value = tci_read_reg16(regs, **tb_ptr); - *tb_ptr += 1; - return value; -} - #if TCG_TARGET_HAS_ext16s_i32 || TCG_TARGET_HAS_ext16s_i64 /* Read indexed register (16 bit signed) from bytecode. */ static int16_t tci_read_r16s(const tcg_target_ulong *regs, @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, *(uint8_t *)(t1 + t2) = t0; break; CASE_32_64(st16) - t0 = tci_read_r16(regs, &tb_ptr); + t0 = tci_read_r(regs, &tb_ptr); t1 = tci_read_r(regs, &tb_ptr); t2 = tci_read_s32(&tb_ptr); *(uint16_t *)(t1 + t2) = t0; @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_ext16u_i32 case INDEX_op_ext16u_i32: t0 = *tb_ptr++; - t1 = tci_read_r16(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (uint16_t)t1); break; #endif #if TCG_TARGET_HAS_bswap16_i32 case INDEX_op_bswap16_i32: t0 = *tb_ptr++; - t1 = tci_read_r16(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, bswap16(t1)); break; #endif @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_ext16u_i64 case INDEX_op_ext16u_i64: t0 = *tb_ptr++; - t1 = tci_read_r16(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (uint16_t)t1); break; #endif #if TCG_TARGET_HAS_ext32s_i64 @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_bswap16_i64 case INDEX_op_bswap16_i64: t0 = *tb_ptr++; - t1 = tci_read_r16(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, bswap16(t1)); break; #endif -- 2.25.1 Use explicit casts for ext16s opcodes. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 26 ++++---------------------- 1 file changed, 4 insertions(+), 22 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ static tcg_target_ulong tci_read_reg(const tcg_target_ulong *regs, TCGReg index) return regs[index]; } -#if TCG_TARGET_HAS_ext16s_i32 || TCG_TARGET_HAS_ext16s_i64 -static int16_t tci_read_reg16s(const tcg_target_ulong *regs, TCGReg index) -{ - return (int16_t)tci_read_reg(regs, index); -} -#endif - #if TCG_TARGET_REG_BITS == 64 static int32_t tci_read_reg32s(const tcg_target_ulong *regs, TCGReg index) { @@ -XXX,XX +XXX,XX @@ tci_read_r(const tcg_target_ulong *regs, const uint8_t **tb_ptr) return value; } -#if TCG_TARGET_HAS_ext16s_i32 || TCG_TARGET_HAS_ext16s_i64 -/* Read indexed register (16 bit signed) from bytecode. */ -static int16_t tci_read_r16s(const tcg_target_ulong *regs, - const uint8_t **tb_ptr) -{ - int16_t value = tci_read_reg16s(regs, **tb_ptr); - *tb_ptr += 1; - return value; -} -#endif - /* Read indexed register (32 bit) from bytecode. */ static uint32_t tci_read_r32(const tcg_target_ulong *regs, const uint8_t **tb_ptr) @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_ext16s_i32 case INDEX_op_ext16s_i32: t0 = *tb_ptr++; - t1 = tci_read_r16s(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (int16_t)t1); break; #endif #if TCG_TARGET_HAS_ext8u_i32 @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_ext16s_i64 case INDEX_op_ext16s_i64: t0 = *tb_ptr++; - t1 = tci_read_r16s(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (int16_t)t1); break; #endif #if TCG_TARGET_HAS_ext16u_i64 -- 2.25.1 Use explicit casts for ext32u opcodes, and allow truncation to happen for other users. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 122 ++++++++++++++++++++++++------------------------------ 1 file changed, 54 insertions(+), 68 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ static int32_t tci_read_reg32s(const tcg_target_ulong *regs, TCGReg index) } #endif -static uint32_t tci_read_reg32(const tcg_target_ulong *regs, TCGReg index) -{ - return (uint32_t)tci_read_reg(regs, index); -} - #if TCG_TARGET_REG_BITS == 64 static uint64_t tci_read_reg64(const tcg_target_ulong *regs, TCGReg index) { @@ -XXX,XX +XXX,XX @@ tci_read_r(const tcg_target_ulong *regs, const uint8_t **tb_ptr) return value; } -/* Read indexed register (32 bit) from bytecode. */ -static uint32_t tci_read_r32(const tcg_target_ulong *regs, - const uint8_t **tb_ptr) -{ - uint32_t value = tci_read_reg32(regs, **tb_ptr); - *tb_ptr += 1; - return value; -} - #if TCG_TARGET_REG_BITS == 32 /* Read two indexed registers (2 * 32 bit) from bytecode. */ static uint64_t tci_read_r64(const tcg_target_ulong *regs, const uint8_t **tb_ptr) { - uint32_t low = tci_read_r32(regs, tb_ptr); - return tci_uint64(tci_read_r32(regs, tb_ptr), low); + uint32_t low = tci_read_r(regs, tb_ptr); + return tci_uint64(tci_read_r(regs, tb_ptr), low); } #elif TCG_TARGET_REG_BITS == 64 /* Read indexed register (32 bit signed) from bytecode. */ @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, continue; case INDEX_op_setcond_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); condition = *tb_ptr++; tci_write_reg(regs, t0, tci_compare32(t1, t2, condition)); break; @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #endif case INDEX_op_mov_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1); break; case INDEX_op_tci_movi_i32: @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, break; case INDEX_op_st_i32: CASE_64(st32) - t0 = tci_read_r32(regs, &tb_ptr); + t0 = tci_read_r(regs, &tb_ptr); t1 = tci_read_r(regs, &tb_ptr); t2 = tci_read_s32(&tb_ptr); *(uint32_t *)(t1 + t2) = t0; @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, case INDEX_op_add_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 + t2); break; case INDEX_op_sub_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 - t2); break; case INDEX_op_mul_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 * t2); break; case INDEX_op_div_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (int32_t)t1 / (int32_t)t2); break; case INDEX_op_divu_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 / t2); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (uint32_t)t1 / (uint32_t)t2); break; case INDEX_op_rem_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (int32_t)t1 % (int32_t)t2); break; case INDEX_op_remu_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 % t2); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (uint32_t)t1 % (uint32_t)t2); break; case INDEX_op_and_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 & t2); break; case INDEX_op_or_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 | t2); break; case INDEX_op_xor_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 ^ t2); break; @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, case INDEX_op_shl_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 << (t2 & 31)); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (uint32_t)t1 << (t2 & 31)); break; case INDEX_op_shr_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 >> (t2 & 31)); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (uint32_t)t1 >> (t2 & 31)); break; case INDEX_op_sar_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); - tci_write_reg(regs, t0, ((int32_t)t1 >> (t2 & 31))); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (int32_t)t1 >> (t2 & 31)); break; #if TCG_TARGET_HAS_rot_i32 case INDEX_op_rotl_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, rol32(t1, t2 & 31)); break; case INDEX_op_rotr_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, ror32(t1, t2 & 31)); break; #endif #if TCG_TARGET_HAS_deposit_i32 case INDEX_op_deposit_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - t2 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tmp16 = *tb_ptr++; tmp8 = *tb_ptr++; tmp32 = (((1 << tmp8) - 1) << tmp16); @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, break; #endif case INDEX_op_brcond_i32: - t0 = tci_read_r32(regs, &tb_ptr); - t1 = tci_read_r32(regs, &tb_ptr); + t0 = tci_read_r(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); condition = *tb_ptr++; label = tci_read_label(&tb_ptr); if (tci_compare32(t0, t1, condition)) { @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, case INDEX_op_mulu2_i32: t0 = *tb_ptr++; t1 = *tb_ptr++; - t2 = tci_read_r32(regs, &tb_ptr); - tmp64 = tci_read_r32(regs, &tb_ptr); - tci_write_reg64(regs, t1, t0, t2 * tmp64); + t2 = tci_read_r(regs, &tb_ptr); + tmp64 = (uint32_t)tci_read_r(regs, &tb_ptr); + tci_write_reg64(regs, t1, t0, (uint32_t)t2 * tmp64); break; #endif /* TCG_TARGET_REG_BITS == 32 */ #if TCG_TARGET_HAS_ext8s_i32 @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_bswap32_i32 case INDEX_op_bswap32_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, bswap32(t1)); break; #endif #if TCG_TARGET_HAS_not_i32 case INDEX_op_not_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, ~t1); break; #endif #if TCG_TARGET_HAS_neg_i32 case INDEX_op_neg_i32: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, -t1); break; #endif @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #endif case INDEX_op_extu_i32_i64: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (uint32_t)t1); break; #if TCG_TARGET_HAS_bswap16_i64 case INDEX_op_bswap16_i64: @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_bswap32_i64 case INDEX_op_bswap32_i64: t0 = *tb_ptr++; - t1 = tci_read_r32(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, bswap32(t1)); break; #endif -- 2.25.1 Use explicit casts for ext32s opcodes. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 20 ++------------------ 1 file changed, 2 insertions(+), 18 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ static tcg_target_ulong tci_read_reg(const tcg_target_ulong *regs, TCGReg index) return regs[index]; } -#if TCG_TARGET_REG_BITS == 64 -static int32_t tci_read_reg32s(const tcg_target_ulong *regs, TCGReg index) -{ - return (int32_t)tci_read_reg(regs, index); -} -#endif - #if TCG_TARGET_REG_BITS == 64 static uint64_t tci_read_reg64(const tcg_target_ulong *regs, TCGReg index) { @@ -XXX,XX +XXX,XX @@ static uint64_t tci_read_r64(const tcg_target_ulong *regs, return tci_uint64(tci_read_r(regs, tb_ptr), low); } #elif TCG_TARGET_REG_BITS == 64 -/* Read indexed register (32 bit signed) from bytecode. */ -static int32_t tci_read_r32s(const tcg_target_ulong *regs, - const uint8_t **tb_ptr) -{ - int32_t value = tci_read_reg32s(regs, **tb_ptr); - *tb_ptr += 1; - return value; -} - /* Read indexed register (64 bit) from bytecode. */ static uint64_t tci_read_r64(const tcg_target_ulong *regs, const uint8_t **tb_ptr) @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #endif case INDEX_op_ext_i32_i64: t0 = *tb_ptr++; - t1 = tci_read_r32s(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); + t1 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, (int32_t)t1); break; #if TCG_TARGET_HAS_ext32u_i64 case INDEX_op_ext32u_i64: -- 2.25.1 In all cases restricted to 64-bit hosts, tcg_read_r is identical. We retain the 64-bit symbol for the single case of INDEX_op_qemu_st_i64. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 93 +++++++++++++++++++++++++------------------------------ 1 file changed, 42 insertions(+), 51 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ static tcg_target_ulong tci_read_reg(const tcg_target_ulong *regs, TCGReg index) return regs[index]; } -#if TCG_TARGET_REG_BITS == 64 -static uint64_t tci_read_reg64(const tcg_target_ulong *regs, TCGReg index) -{ - return tci_read_reg(regs, index); -} -#endif - static void tci_write_reg(tcg_target_ulong *regs, TCGReg index, tcg_target_ulong value) { @@ -XXX,XX +XXX,XX @@ static uint64_t tci_read_r64(const tcg_target_ulong *regs, static uint64_t tci_read_r64(const tcg_target_ulong *regs, const uint8_t **tb_ptr) { - uint64_t value = tci_read_reg64(regs, **tb_ptr); - *tb_ptr += 1; - return value; + return tci_read_r(regs, tb_ptr); } #endif @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #elif TCG_TARGET_REG_BITS == 64 case INDEX_op_setcond_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); condition = *tb_ptr++; tci_write_reg(regs, t0, tci_compare64(t1, t2, condition)); break; @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_REG_BITS == 64 case INDEX_op_mov_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1); break; case INDEX_op_tci_movi_i64: @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, tci_write_reg(regs, t0, *(uint64_t *)(t1 + t2)); break; case INDEX_op_st_i64: - t0 = tci_read_r64(regs, &tb_ptr); + t0 = tci_read_r(regs, &tb_ptr); t1 = tci_read_r(regs, &tb_ptr); t2 = tci_read_s32(&tb_ptr); *(uint64_t *)(t1 + t2) = t0; @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, case INDEX_op_add_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 + t2); break; case INDEX_op_sub_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 - t2); break; case INDEX_op_mul_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 * t2); break; case INDEX_op_div_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (int64_t)t1 / (int64_t)t2); break; case INDEX_op_divu_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (uint64_t)t1 / (uint64_t)t2); break; case INDEX_op_rem_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (int64_t)t1 % (int64_t)t2); break; case INDEX_op_remu_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (uint64_t)t1 % (uint64_t)t2); break; case INDEX_op_and_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 & t2); break; case INDEX_op_or_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 | t2); break; case INDEX_op_xor_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 ^ t2); break; @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, case INDEX_op_shl_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 << (t2 & 63)); break; case INDEX_op_shr_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 >> (t2 & 63)); break; case INDEX_op_sar_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, ((int64_t)t1 >> (t2 & 63))); break; #if TCG_TARGET_HAS_rot_i64 case INDEX_op_rotl_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, rol64(t1, t2 & 63)); break; case INDEX_op_rotr_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, ror64(t1, t2 & 63)); break; #endif #if TCG_TARGET_HAS_deposit_i64 case INDEX_op_deposit_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); - t2 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); tmp16 = *tb_ptr++; tmp8 = *tb_ptr++; tmp64 = (((1ULL << tmp8) - 1) << tmp16); @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, break; #endif case INDEX_op_brcond_i64: - t0 = tci_read_r64(regs, &tb_ptr); - t1 = tci_read_r64(regs, &tb_ptr); + t0 = tci_read_r(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); condition = *tb_ptr++; label = tci_read_label(&tb_ptr); if (tci_compare64(t0, t1, condition)) { @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, #if TCG_TARGET_HAS_bswap64_i64 case INDEX_op_bswap64_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, bswap64(t1)); break; #endif #if TCG_TARGET_HAS_not_i64 case INDEX_op_not_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, ~t1); break; #endif #if TCG_TARGET_HAS_neg_i64 case INDEX_op_neg_i64: t0 = *tb_ptr++; - t1 = tci_read_r64(regs, &tb_ptr); + t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, -t1); break; #endif -- 2.25.1 This includes add, sub, mul, and, or, xor. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 83 +++++++++++++++++-------------------------------------- 1 file changed, 25 insertions(+), 58 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, *(uint32_t *)(t1 + t2) = t0; break; - /* Arithmetic operations (32 bit). */ + /* Arithmetic operations (mixed 32/64 bit). */ - case INDEX_op_add_i32: + CASE_32_64(add) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 + t2); break; - case INDEX_op_sub_i32: + CASE_32_64(sub) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 - t2); break; - case INDEX_op_mul_i32: + CASE_32_64(mul) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1 * t2); break; + CASE_32_64(and) + t0 = *tb_ptr++; + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, t1 & t2); + break; + CASE_32_64(or) + t0 = *tb_ptr++; + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, t1 | t2); + break; + CASE_32_64(xor) + t0 = *tb_ptr++; + t1 = tci_read_r(regs, &tb_ptr); + t2 = tci_read_r(regs, &tb_ptr); + tci_write_reg(regs, t0, t1 ^ t2); + break; + + /* Arithmetic operations (32 bit). */ + case INDEX_op_div_i32: t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (uint32_t)t1 % (uint32_t)t2); break; - case INDEX_op_and_i32: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - t2 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 & t2); - break; - case INDEX_op_or_i32: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - t2 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 | t2); - break; - case INDEX_op_xor_i32: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - t2 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 ^ t2); - break; /* Shift/rotate operations (32 bit). */ @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, /* Arithmetic operations (64 bit). */ - case INDEX_op_add_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - t2 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 + t2); - break; - case INDEX_op_sub_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - t2 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 - t2); - break; - case INDEX_op_mul_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - t2 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 * t2); - break; case INDEX_op_div_i64: t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, t2 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (uint64_t)t1 % (uint64_t)t2); break; - case INDEX_op_and_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - t2 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 & t2); - break; - case INDEX_op_or_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - t2 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 | t2); - break; - case INDEX_op_xor_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - t2 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1 ^ t2); - break; /* Shift/rotate operations (64 bit). */ -- 2.25.1 This includes ext8s, ext8u, ext16s, ext16u. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 44 ++++++++------------------------------------ 1 file changed, 8 insertions(+), 36 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, tci_write_reg64(regs, t1, t0, (uint32_t)t2 * tmp64); break; #endif /* TCG_TARGET_REG_BITS == 32 */ -#if TCG_TARGET_HAS_ext8s_i32 - case INDEX_op_ext8s_i32: +#if TCG_TARGET_HAS_ext8s_i32 || TCG_TARGET_HAS_ext8s_i64 + CASE_32_64(ext8s) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (int8_t)t1); break; #endif -#if TCG_TARGET_HAS_ext16s_i32 - case INDEX_op_ext16s_i32: +#if TCG_TARGET_HAS_ext16s_i32 || TCG_TARGET_HAS_ext16s_i64 + CASE_32_64(ext16s) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (int16_t)t1); break; #endif -#if TCG_TARGET_HAS_ext8u_i32 - case INDEX_op_ext8u_i32: +#if TCG_TARGET_HAS_ext8u_i32 || TCG_TARGET_HAS_ext8u_i64 + CASE_32_64(ext8u) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (uint8_t)t1); break; #endif -#if TCG_TARGET_HAS_ext16u_i32 - case INDEX_op_ext16u_i32: +#if TCG_TARGET_HAS_ext16u_i32 || TCG_TARGET_HAS_ext16u_i64 + CASE_32_64(ext16u) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (uint16_t)t1); @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, continue; } break; -#if TCG_TARGET_HAS_ext8u_i64 - case INDEX_op_ext8u_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, (uint8_t)t1); - break; -#endif -#if TCG_TARGET_HAS_ext8s_i64 - case INDEX_op_ext8s_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, (int8_t)t1); - break; -#endif -#if TCG_TARGET_HAS_ext16s_i64 - case INDEX_op_ext16s_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, (int16_t)t1); - break; -#endif -#if TCG_TARGET_HAS_ext16u_i64 - case INDEX_op_ext16u_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, (uint16_t)t1); - break; -#endif #if TCG_TARGET_HAS_ext32s_i64 case INDEX_op_ext32s_i64: #endif -- 2.25.1 This includes bswap16 and bswap32. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 22 ++++------------------ 1 file changed, 4 insertions(+), 18 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, tci_write_reg(regs, t0, (uint16_t)t1); break; #endif -#if TCG_TARGET_HAS_bswap16_i32 - case INDEX_op_bswap16_i32: +#if TCG_TARGET_HAS_bswap16_i32 || TCG_TARGET_HAS_bswap16_i64 + CASE_32_64(bswap16) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, bswap16(t1)); break; #endif -#if TCG_TARGET_HAS_bswap32_i32 - case INDEX_op_bswap32_i32: +#if TCG_TARGET_HAS_bswap32_i32 || TCG_TARGET_HAS_bswap32_i64 + CASE_32_64(bswap32) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, bswap32(t1)); @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, (uint32_t)t1); break; -#if TCG_TARGET_HAS_bswap16_i64 - case INDEX_op_bswap16_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, bswap16(t1)); - break; -#endif -#if TCG_TARGET_HAS_bswap32_i64 - case INDEX_op_bswap32_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, bswap32(t1)); - break; -#endif #if TCG_TARGET_HAS_bswap64_i64 case INDEX_op_bswap64_i64: t0 = *tb_ptr++; -- 2.25.1 Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tci.c | 29 +++++------------------------ 1 file changed, 5 insertions(+), 24 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, tci_write_reg(regs, t0, tci_compare64(t1, t2, condition)); break; #endif - case INDEX_op_mov_i32: + CASE_32_64(mov) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, t1); @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, tci_write_reg(regs, t0, bswap32(t1)); break; #endif -#if TCG_TARGET_HAS_not_i32 - case INDEX_op_not_i32: +#if TCG_TARGET_HAS_not_i32 || TCG_TARGET_HAS_not_i64 + CASE_32_64(not) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, ~t1); break; #endif -#if TCG_TARGET_HAS_neg_i32 - case INDEX_op_neg_i32: +#if TCG_TARGET_HAS_neg_i32 || TCG_TARGET_HAS_neg_i64 + CASE_32_64(neg) t0 = *tb_ptr++; t1 = tci_read_r(regs, &tb_ptr); tci_write_reg(regs, t0, -t1); break; #endif #if TCG_TARGET_REG_BITS == 64 - case INDEX_op_mov_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, t1); - break; case INDEX_op_tci_movi_i64: t0 = *tb_ptr++; t1 = tci_read_i64(&tb_ptr); @@ -XXX,XX +XXX,XX @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, tci_write_reg(regs, t0, bswap64(t1)); break; #endif -#if TCG_TARGET_HAS_not_i64 - case INDEX_op_not_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, ~t1); - break; -#endif -#if TCG_TARGET_HAS_neg_i64 - case INDEX_op_neg_i64: - t0 = *tb_ptr++; - t1 = tci_read_r(regs, &tb_ptr); - tci_write_reg(regs, t0, -t1); - break; -#endif #endif /* TCG_TARGET_REG_BITS == 64 */ /* QEMU specific operations. */ -- 2.25.1 From: Alex Bennée <alex.bennee@linaro.org> Having a function return either and valid TB and some system state seems excessive. It will make the subsequent re-factoring easier if we lookup the current state where we are. Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Message-Id: <20210224165811.11567-2-alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/tb-lookup.h | 18 ++++++++---------- accel/tcg/cpu-exec.c | 10 ++++++++-- accel/tcg/tcg-runtime.c | 4 +++- 3 files changed, 19 insertions(+), 13 deletions(-) diff --git a/include/exec/tb-lookup.h b/include/exec/tb-lookup.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/tb-lookup.h +++ b/include/exec/tb-lookup.h @@ -XXX,XX +XXX,XX @@ #include "exec/tb-hash.h" /* Might cause an exception, so have a longjmp destination ready */ -static inline TranslationBlock * -tb_lookup__cpu_state(CPUState *cpu, target_ulong *pc, target_ulong *cs_base, - uint32_t *flags, uint32_t cf_mask) +static inline TranslationBlock * tb_lookup(CPUState *cpu, + target_ulong pc, target_ulong cs_base, + uint32_t flags, uint32_t cf_mask) { - CPUArchState *env = (CPUArchState *)cpu->env_ptr; TranslationBlock *tb; uint32_t hash; - cpu_get_tb_cpu_state(env, pc, cs_base, flags); - hash = tb_jmp_cache_hash_func(*pc); + hash = tb_jmp_cache_hash_func(pc); tb = qatomic_rcu_read(&cpu->tb_jmp_cache[hash]); cf_mask &= ~CF_CLUSTER_MASK; cf_mask |= cpu->cluster_index << CF_CLUSTER_SHIFT; if (likely(tb && - tb->pc == *pc && - tb->cs_base == *cs_base && - tb->flags == *flags && + tb->pc == pc && + tb->cs_base == cs_base && + tb->flags == flags && tb->trace_vcpu_dstate == *cpu->trace_dstate && (tb_cflags(tb) & (CF_HASH_MASK | CF_INVALID)) == cf_mask)) { return tb; } - tb = tb_htable_lookup(cpu, *pc, *cs_base, *flags, cf_mask); + tb = tb_htable_lookup(cpu, pc, cs_base, flags, cf_mask); if (tb == NULL) { return NULL; } diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ static void cpu_exec_exit(CPUState *cpu) void cpu_exec_step_atomic(CPUState *cpu) { + CPUArchState *env = (CPUArchState *)cpu->env_ptr; TranslationBlock *tb; target_ulong cs_base, pc; uint32_t flags; @@ -XXX,XX +XXX,XX @@ void cpu_exec_step_atomic(CPUState *cpu) g_assert(!cpu->running); cpu->running = true; - tb = tb_lookup__cpu_state(cpu, &pc, &cs_base, &flags, cf_mask); + cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); + tb = tb_lookup(cpu, pc, cs_base, flags, cf_mask); + if (tb == NULL) { mmap_lock(); tb = tb_gen_code(cpu, pc, cs_base, flags, cflags); @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_find(CPUState *cpu, TranslationBlock *last_tb, int tb_exit, uint32_t cf_mask) { + CPUArchState *env = (CPUArchState *)cpu->env_ptr; TranslationBlock *tb; target_ulong cs_base, pc; uint32_t flags; - tb = tb_lookup__cpu_state(cpu, &pc, &cs_base, &flags, cf_mask); + cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); + + tb = tb_lookup(cpu, pc, cs_base, flags, cf_mask); if (tb == NULL) { mmap_lock(); tb = tb_gen_code(cpu, pc, cs_base, flags, cf_mask); diff --git a/accel/tcg/tcg-runtime.c b/accel/tcg/tcg-runtime.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tcg-runtime.c +++ b/accel/tcg/tcg-runtime.c @@ -XXX,XX +XXX,XX @@ const void *HELPER(lookup_tb_ptr)(CPUArchState *env) target_ulong cs_base, pc; uint32_t flags; - tb = tb_lookup__cpu_state(cpu, &pc, &cs_base, &flags, curr_cflags()); + cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); + + tb = tb_lookup(cpu, pc, cs_base, flags, curr_cflags()); if (tb == NULL) { return tcg_code_gen_epilogue; } -- 2.25.1 From: Alex Bennée <alex.bennee@linaro.org> There is nothing special about this compile flag that doesn't mean we can't just compute it with curr_cflags() which we should be using when building a new set. Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Message-Id: <20210224165811.11567-3-alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/exec-all.h | 8 +++++--- include/exec/tb-lookup.h | 3 --- accel/tcg/cpu-exec.c | 9 ++++----- accel/tcg/tcg-runtime.c | 2 +- accel/tcg/translate-all.c | 6 +++--- softmmu/physmem.c | 2 +- 6 files changed, 14 insertions(+), 16 deletions(-) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ static inline uint32_t tb_cflags(const TranslationBlock *tb) } /* current cflags for hashing/comparison */ -static inline uint32_t curr_cflags(void) +static inline uint32_t curr_cflags(CPUState *cpu) { - return (parallel_cpus ? CF_PARALLEL : 0) - | (icount_enabled() ? CF_USE_ICOUNT : 0); + uint32_t cflags = deposit32(0, CF_CLUSTER_SHIFT, 8, cpu->cluster_index); + cflags |= parallel_cpus ? CF_PARALLEL : 0; + cflags |= icount_enabled() ? CF_USE_ICOUNT : 0; + return cflags; } /* TranslationBlock invalidate API */ diff --git a/include/exec/tb-lookup.h b/include/exec/tb-lookup.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/tb-lookup.h +++ b/include/exec/tb-lookup.h @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock * tb_lookup(CPUState *cpu, hash = tb_jmp_cache_hash_func(pc); tb = qatomic_rcu_read(&cpu->tb_jmp_cache[hash]); - cf_mask &= ~CF_CLUSTER_MASK; - cf_mask |= cpu->cluster_index << CF_CLUSTER_SHIFT; - if (likely(tb && tb->pc == pc && tb->cs_base == cs_base && diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ void cpu_exec_step_atomic(CPUState *cpu) TranslationBlock *tb; target_ulong cs_base, pc; uint32_t flags; - uint32_t cflags = 1; - uint32_t cf_mask = cflags & CF_HASH_MASK; + uint32_t cflags = (curr_cflags(cpu) & ~CF_PARALLEL) | 1; int tb_exit; if (sigsetjmp(cpu->jmp_env, 0) == 0) { @@ -XXX,XX +XXX,XX @@ void cpu_exec_step_atomic(CPUState *cpu) cpu->running = true; cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); - tb = tb_lookup(cpu, pc, cs_base, flags, cf_mask); + tb = tb_lookup(cpu, pc, cs_base, flags, cflags); if (tb == NULL) { mmap_lock(); @@ -XXX,XX +XXX,XX @@ static inline bool cpu_handle_exception(CPUState *cpu, int *ret) if (replay_has_exception() && cpu_neg(cpu)->icount_decr.u16.low + cpu->icount_extra == 0) { /* Execute just one insn to trigger exception pending in the log */ - cpu->cflags_next_tb = (curr_cflags() & ~CF_USE_ICOUNT) | 1; + cpu->cflags_next_tb = (curr_cflags(cpu) & ~CF_USE_ICOUNT) | 1; } #endif return false; @@ -XXX,XX +XXX,XX @@ int cpu_exec(CPUState *cpu) have CF_INVALID set, -1 is a convenient invalid value that does not require tcg headers for cpu_common_reset. */ if (cflags == -1) { - cflags = curr_cflags(); + cflags = curr_cflags(cpu); } else { cpu->cflags_next_tb = -1; } diff --git a/accel/tcg/tcg-runtime.c b/accel/tcg/tcg-runtime.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tcg-runtime.c +++ b/accel/tcg/tcg-runtime.c @@ -XXX,XX +XXX,XX @@ const void *HELPER(lookup_tb_ptr)(CPUArchState *env) cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); - tb = tb_lookup(cpu, pc, cs_base, flags, curr_cflags()); + tb = tb_lookup(cpu, pc, cs_base, flags, curr_cflags(cpu)); if (tb == NULL) { return tcg_code_gen_epilogue; } diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ tb_invalidate_phys_page_range__locked(struct page_collection *pages, if (current_tb_modified) { page_collection_unlock(pages); /* Force execution of one insn next time. */ - cpu->cflags_next_tb = 1 | curr_cflags(); + cpu->cflags_next_tb = 1 | curr_cflags(cpu); mmap_unlock(); cpu_loop_exit_noexc(cpu); } @@ -XXX,XX +XXX,XX @@ static bool tb_invalidate_phys_page(tb_page_addr_t addr, uintptr_t pc) #ifdef TARGET_HAS_PRECISE_SMC if (current_tb_modified) { /* Force execution of one insn next time. */ - cpu->cflags_next_tb = 1 | curr_cflags(); + cpu->cflags_next_tb = 1 | curr_cflags(cpu); return true; } #endif @@ -XXX,XX +XXX,XX @@ void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr) * operations only (which execute after completion) so we don't * double instrument the instruction. */ - cpu->cflags_next_tb = curr_cflags() | CF_MEMI_ONLY | CF_LAST_IO | n; + cpu->cflags_next_tb = curr_cflags(cpu) | CF_MEMI_ONLY | CF_LAST_IO | n; qemu_log_mask_and_addr(CPU_LOG_EXEC, tb->pc, "cpu_io_recompile: rewound execution of TB to " diff --git a/softmmu/physmem.c b/softmmu/physmem.c index XXXXXXX..XXXXXXX 100644 --- a/softmmu/physmem.c +++ b/softmmu/physmem.c @@ -XXX,XX +XXX,XX @@ void cpu_check_watchpoint(CPUState *cpu, vaddr addr, vaddr len, cpu_loop_exit_restore(cpu, ra); } else { /* Force execution of one insn next time. */ - cpu->cflags_next_tb = 1 | curr_cflags(); + cpu->cflags_next_tb = 1 | curr_cflags(cpu); mmap_unlock(); if (ra) { cpu_restore_state(cpu, ra, true); -- 2.25.1 From: Alex Bennée <alex.bennee@linaro.org> We don't really deal in cf_mask most of the time. The one time it's relevant is when we want to remove an invalidated TB from the QHT lookup. Everywhere else we should be looking up things without CF_INVALID set. Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Message-Id: <20210224165811.11567-4-alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/exec-all.h | 4 +--- include/exec/tb-lookup.h | 9 ++++++--- accel/tcg/cpu-exec.c | 16 ++++++++-------- accel/tcg/tcg-runtime.c | 2 +- accel/tcg/translate-all.c | 8 +++++--- 5 files changed, 21 insertions(+), 18 deletions(-) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ struct TranslationBlock { #define CF_PARALLEL 0x00080000 /* Generate code for a parallel context */ #define CF_CLUSTER_MASK 0xff000000 /* Top 8 bits are cluster ID */ #define CF_CLUSTER_SHIFT 24 -/* cflags' mask for hashing/comparison, basically ignore CF_INVALID */ -#define CF_HASH_MASK (~CF_INVALID) /* Per-vCPU dynamic tracing state used to generate this TB */ uint32_t trace_vcpu_dstate; @@ -XXX,XX +XXX,XX @@ void tb_flush(CPUState *cpu); void tb_phys_invalidate(TranslationBlock *tb, tb_page_addr_t page_addr); TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, target_ulong cs_base, uint32_t flags, - uint32_t cf_mask); + uint32_t cflags); void tb_set_jmp_target(TranslationBlock *tb, int n, uintptr_t addr); /* GETPC is the true target of the return instruction that we'll execute. */ diff --git a/include/exec/tb-lookup.h b/include/exec/tb-lookup.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/tb-lookup.h +++ b/include/exec/tb-lookup.h @@ -XXX,XX +XXX,XX @@ /* Might cause an exception, so have a longjmp destination ready */ static inline TranslationBlock * tb_lookup(CPUState *cpu, target_ulong pc, target_ulong cs_base, - uint32_t flags, uint32_t cf_mask) + uint32_t flags, uint32_t cflags) { TranslationBlock *tb; uint32_t hash; + /* we should never be trying to look up an INVALID tb */ + tcg_debug_assert(!(cflags & CF_INVALID)); + hash = tb_jmp_cache_hash_func(pc); tb = qatomic_rcu_read(&cpu->tb_jmp_cache[hash]); @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock * tb_lookup(CPUState *cpu, tb->cs_base == cs_base && tb->flags == flags && tb->trace_vcpu_dstate == *cpu->trace_dstate && - (tb_cflags(tb) & (CF_HASH_MASK | CF_INVALID)) == cf_mask)) { + tb_cflags(tb) == cflags)) { return tb; } - tb = tb_htable_lookup(cpu, pc, cs_base, flags, cf_mask); + tb = tb_htable_lookup(cpu, pc, cs_base, flags, cflags); if (tb == NULL) { return NULL; } diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ struct tb_desc { CPUArchState *env; tb_page_addr_t phys_page1; uint32_t flags; - uint32_t cf_mask; + uint32_t cflags; uint32_t trace_vcpu_dstate; }; @@ -XXX,XX +XXX,XX @@ static bool tb_lookup_cmp(const void *p, const void *d) tb->cs_base == desc->cs_base && tb->flags == desc->flags && tb->trace_vcpu_dstate == desc->trace_vcpu_dstate && - (tb_cflags(tb) & (CF_HASH_MASK | CF_INVALID)) == desc->cf_mask) { + tb_cflags(tb) == desc->cflags) { /* check next page if needed */ if (tb->page_addr[1] == -1) { return true; @@ -XXX,XX +XXX,XX @@ static bool tb_lookup_cmp(const void *p, const void *d) TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, target_ulong cs_base, uint32_t flags, - uint32_t cf_mask) + uint32_t cflags) { tb_page_addr_t phys_pc; struct tb_desc desc; @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, desc.env = (CPUArchState *)cpu->env_ptr; desc.cs_base = cs_base; desc.flags = flags; - desc.cf_mask = cf_mask; + desc.cflags = cflags; desc.trace_vcpu_dstate = *cpu->trace_dstate; desc.pc = pc; phys_pc = get_page_addr_code(desc.env, pc); @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, return NULL; } desc.phys_page1 = phys_pc & TARGET_PAGE_MASK; - h = tb_hash_func(phys_pc, pc, flags, cf_mask, *cpu->trace_dstate); + h = tb_hash_func(phys_pc, pc, flags, cflags, *cpu->trace_dstate); return qht_lookup_custom(&tb_ctx.htable, &desc, h, tb_lookup_cmp); } @@ -XXX,XX +XXX,XX @@ static inline void tb_add_jump(TranslationBlock *tb, int n, static inline TranslationBlock *tb_find(CPUState *cpu, TranslationBlock *last_tb, - int tb_exit, uint32_t cf_mask) + int tb_exit, uint32_t cflags) { CPUArchState *env = (CPUArchState *)cpu->env_ptr; TranslationBlock *tb; @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_find(CPUState *cpu, cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); - tb = tb_lookup(cpu, pc, cs_base, flags, cf_mask); + tb = tb_lookup(cpu, pc, cs_base, flags, cflags); if (tb == NULL) { mmap_lock(); - tb = tb_gen_code(cpu, pc, cs_base, flags, cf_mask); + tb = tb_gen_code(cpu, pc, cs_base, flags, cflags); mmap_unlock(); /* We add the TB in the virtual pc hash table for the fast lookup */ qatomic_set(&cpu->tb_jmp_cache[tb_jmp_cache_hash_func(pc)], tb); diff --git a/accel/tcg/tcg-runtime.c b/accel/tcg/tcg-runtime.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tcg-runtime.c +++ b/accel/tcg/tcg-runtime.c @@ -XXX,XX +XXX,XX @@ #include "exec/helper-proto.h" #include "exec/cpu_ldst.h" #include "exec/exec-all.h" -#include "exec/tb-lookup.h" #include "disas/disas.h" #include "exec/log.h" #include "tcg/tcg.h" +#include "exec/tb-lookup.h" /* 32-bit helpers */ diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ static bool tb_cmp(const void *ap, const void *bp) return a->pc == b->pc && a->cs_base == b->cs_base && a->flags == b->flags && - (tb_cflags(a) & CF_HASH_MASK) == (tb_cflags(b) & CF_HASH_MASK) && + (tb_cflags(a) & ~CF_INVALID) == (tb_cflags(b) & ~CF_INVALID) && a->trace_vcpu_dstate == b->trace_vcpu_dstate && a->page_addr[0] == b->page_addr[0] && a->page_addr[1] == b->page_addr[1]; @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) PageDesc *p; uint32_t h; tb_page_addr_t phys_pc; + uint32_t orig_cflags = tb_cflags(tb); assert_memory_lock(); @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) /* remove the TB from the hash list */ phys_pc = tb->page_addr[0] + (tb->pc & ~TARGET_PAGE_MASK); - h = tb_hash_func(phys_pc, tb->pc, tb->flags, tb_cflags(tb) & CF_HASH_MASK, + h = tb_hash_func(phys_pc, tb->pc, tb->flags, orig_cflags, tb->trace_vcpu_dstate); if (!qht_remove(&tb_ctx.htable, tb, h)) { return; @@ -XXX,XX +XXX,XX @@ tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, uint32_t h; assert_memory_lock(); + tcg_debug_assert(!(tb->cflags & CF_INVALID)); /* * Add the TB to the page list, acquiring first the pages's locks. @@ -XXX,XX +XXX,XX @@ tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, } /* add in the hash table */ - h = tb_hash_func(phys_pc, tb->pc, tb->flags, tb->cflags & CF_HASH_MASK, + h = tb_hash_func(phys_pc, tb->pc, tb->flags, tb->cflags, tb->trace_vcpu_dstate); qht_insert(&tb_ctx.htable, tb, h, &existing_tb); -- 2.25.1 From: Alex Bennée <alex.bennee@linaro.org> Lets make sure all the flags we compare when looking up blocks are together in the same place. Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Message-Id: <20210224165811.11567-5-alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/exec-all.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ struct TranslationBlock { target_ulong pc; /* simulated PC corresponding to this block (EIP + CS base) */ target_ulong cs_base; /* CS base for this block */ uint32_t flags; /* flags defining in which context the code was generated */ - uint16_t size; /* size of target code for this block (1 <= - size <= TARGET_PAGE_SIZE) */ - uint16_t icount; uint32_t cflags; /* compile flags */ #define CF_COUNT_MASK 0x00007fff #define CF_LAST_IO 0x00008000 /* Last insn may be an IO access. */ @@ -XXX,XX +XXX,XX @@ struct TranslationBlock { /* Per-vCPU dynamic tracing state used to generate this TB */ uint32_t trace_vcpu_dstate; + /* Above fields used for comparing */ + uint16_t size; /* size of target code for this block (1 <= + size <= TARGET_PAGE_SIZE) */ + uint16_t icount; + struct tb_tc tc; /* first and second physical page containing code. The lower bit -- 2.25.1 The primary motivation is to remove a dozen insns along the fast-path in tb_lookup. As a byproduct, this allows us to completely remove parallel_cpus. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/tcg-accel-ops.h | 1 + include/exec/exec-all.h | 7 +------ include/hw/core/cpu.h | 2 ++ accel/tcg/cpu-exec.c | 3 --- accel/tcg/tcg-accel-ops-mttcg.c | 3 +-- accel/tcg/tcg-accel-ops-rr.c | 2 +- accel/tcg/tcg-accel-ops.c | 8 ++++++++ accel/tcg/translate-all.c | 4 ---- linux-user/main.c | 1 + linux-user/sh4/signal.c | 8 +++++--- linux-user/syscall.c | 18 ++++++++++-------- 11 files changed, 30 insertions(+), 27 deletions(-) diff --git a/accel/tcg/tcg-accel-ops.h b/accel/tcg/tcg-accel-ops.h index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tcg-accel-ops.h +++ b/accel/tcg/tcg-accel-ops.h @@ -XXX,XX +XXX,XX @@ void tcg_cpus_destroy(CPUState *cpu); int tcg_cpus_exec(CPUState *cpu); void tcg_handle_interrupt(CPUState *cpu, int mask); +void tcg_cpu_init_cflags(CPUState *cpu, bool parallel); #endif /* TCG_CPUS_H */ diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ struct TranslationBlock { uintptr_t jmp_dest[2]; }; -extern bool parallel_cpus; - /* Hide the qatomic_read to make code a little easier on the eyes */ static inline uint32_t tb_cflags(const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static inline uint32_t tb_cflags(const TranslationBlock *tb) /* current cflags for hashing/comparison */ static inline uint32_t curr_cflags(CPUState *cpu) { - uint32_t cflags = deposit32(0, CF_CLUSTER_SHIFT, 8, cpu->cluster_index); - cflags |= parallel_cpus ? CF_PARALLEL : 0; - cflags |= icount_enabled() ? CF_USE_ICOUNT : 0; - return cflags; + return cpu->tcg_cflags; } /* TranslationBlock invalidate API */ diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/core/cpu.h +++ b/include/hw/core/cpu.h @@ -XXX,XX +XXX,XX @@ struct qemu_work_item; * to a cluster this will be UNASSIGNED_CLUSTER_INDEX; otherwise it will * be the same as the cluster-id property of the CPU object's TYPE_CPU_CLUSTER * QOM parent. + * @tcg_cflags: Pre-computed cflags for this cpu. * @nr_cores: Number of cores within this CPU package. * @nr_threads: Number of threads within this CPU. * @running: #true if CPU is currently running (lockless). @@ -XXX,XX +XXX,XX @@ struct CPUState { /* TODO Move common fields from CPUArchState here. */ int cpu_index; int cluster_index; + uint32_t tcg_cflags; uint32_t halted; uint32_t can_do_io; int32_t exception_index; diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ void cpu_exec_step_atomic(CPUState *cpu) mmap_unlock(); } - /* Since we got here, we know that parallel_cpus must be true. */ - parallel_cpus = false; cpu_exec_enter(cpu); /* execute the generated code */ trace_exec_tb(tb, pc); @@ -XXX,XX +XXX,XX @@ void cpu_exec_step_atomic(CPUState *cpu) * the execution. */ g_assert(cpu_in_exclusive_context(cpu)); - parallel_cpus = true; cpu->running = false; end_exclusive(); } diff --git a/accel/tcg/tcg-accel-ops-mttcg.c b/accel/tcg/tcg-accel-ops-mttcg.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tcg-accel-ops-mttcg.c +++ b/accel/tcg/tcg-accel-ops-mttcg.c @@ -XXX,XX +XXX,XX @@ void mttcg_start_vcpu_thread(CPUState *cpu) char thread_name[VCPU_THREAD_NAME_SIZE]; g_assert(tcg_enabled()); - - parallel_cpus = (current_machine->smp.max_cpus > 1); + tcg_cpu_init_cflags(cpu, current_machine->smp.max_cpus > 1); cpu->thread = g_malloc0(sizeof(QemuThread)); cpu->halt_cond = g_malloc0(sizeof(QemuCond)); diff --git a/accel/tcg/tcg-accel-ops-rr.c b/accel/tcg/tcg-accel-ops-rr.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tcg-accel-ops-rr.c +++ b/accel/tcg/tcg-accel-ops-rr.c @@ -XXX,XX +XXX,XX @@ void rr_start_vcpu_thread(CPUState *cpu) static QemuThread *single_tcg_cpu_thread; g_assert(tcg_enabled()); - parallel_cpus = false; + tcg_cpu_init_cflags(cpu, false); if (!single_tcg_cpu_thread) { cpu->thread = g_malloc0(sizeof(QemuThread)); diff --git a/accel/tcg/tcg-accel-ops.c b/accel/tcg/tcg-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tcg-accel-ops.c +++ b/accel/tcg/tcg-accel-ops.c @@ -XXX,XX +XXX,XX @@ /* common functionality among all TCG variants */ +void tcg_cpu_init_cflags(CPUState *cpu, bool parallel) +{ + uint32_t cflags = cpu->cluster_index << CF_CLUSTER_SHIFT; + cflags |= parallel ? CF_PARALLEL : 0; + cflags |= icount_enabled() ? CF_USE_ICOUNT : 0; + cpu->tcg_cflags = cflags; +} + void tcg_cpus_destroy(CPUState *cpu) { cpu_thread_signal_destroyed(cpu); diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ static void *l1_map[V_L1_MAX_SIZE]; TCGContext tcg_init_ctx; __thread TCGContext *tcg_ctx; TBContext tb_ctx; -bool parallel_cpus; static void page_table_config_init(void) { @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_gen_code(CPUState *cpu, cflags = (cflags & ~CF_COUNT_MASK) | 1; } - cflags &= ~CF_CLUSTER_MASK; - cflags |= cpu->cluster_index << CF_CLUSTER_SHIFT; - max_insns = cflags & CF_COUNT_MASK; if (max_insns == 0) { max_insns = CF_COUNT_MASK; diff --git a/linux-user/main.c b/linux-user/main.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/main.c +++ b/linux-user/main.c @@ -XXX,XX +XXX,XX @@ CPUArchState *cpu_copy(CPUArchState *env) /* Reset non arch specific state */ cpu_reset(new_cpu); + new_cpu->tcg_cflags = cpu->tcg_cflags; memcpy(new_env, env, sizeof(CPUArchState)); /* Clone all break/watchpoints. diff --git a/linux-user/sh4/signal.c b/linux-user/sh4/signal.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/sh4/signal.c +++ b/linux-user/sh4/signal.c @@ -XXX,XX +XXX,XX @@ static abi_ulong get_sigframe(struct target_sigaction *ka, return (sp - frame_size) & -8ul; } -/* Notice when we're in the middle of a gUSA region and reset. - Note that this will only occur for !parallel_cpus, as we will - translate such sequences differently in a parallel context. */ +/* + * Notice when we're in the middle of a gUSA region and reset. + * Note that this will only occur when #CF_PARALLEL is unset, as we + * will translate such sequences differently in a parallel context. + */ static void unwind_gusa(CPUSH4State *regs) { /* If the stack pointer is sufficiently negative, and we haven't diff --git a/linux-user/syscall.c b/linux-user/syscall.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/syscall.c +++ b/linux-user/syscall.c @@ -XXX,XX +XXX,XX @@ static int do_fork(CPUArchState *env, unsigned int flags, abi_ulong newsp, /* Grab a mutex so that thread setup appears atomic. */ pthread_mutex_lock(&clone_lock); + /* + * If this is our first additional thread, we need to ensure we + * generate code for parallel execution and flush old translations. + * Do this now so that the copy gets CF_PARALLEL too. + */ + if (!(cpu->tcg_cflags & CF_PARALLEL)) { + cpu->tcg_cflags |= CF_PARALLEL; + tb_flush(cpu); + } + /* we create a new CPU instance. */ new_env = cpu_copy(env); /* Init regs that differ from the parent. */ @@ -XXX,XX +XXX,XX @@ static int do_fork(CPUArchState *env, unsigned int flags, abi_ulong newsp, sigprocmask(SIG_BLOCK, &sigmask, &info.sigmask); cpu->random_seed = qemu_guest_random_seed_thread_part1(); - /* If this is our first additional thread, we need to ensure we - * generate code for parallel execution and flush old translations. - */ - if (!parallel_cpus) { - parallel_cpus = true; - tb_flush(cpu); - } - ret = pthread_create(&info.thread, &attr, clone_func, &info); /* TODO: Free new CPU state if thread creation failed. */ -- 2.25.1 TCG patch queue, plus one target/sh4 patch that Yoshinori Sato asked me to process. r~ The following changes since commit efbf38d73e5dcc4d5f8b98c6e7a12be1f3b91745: Merge tag 'for-upstream' of git://repo.or.cz/qemu/kevin into staging (2022-10-03 15:06:07 -0400) are available in the Git repository at: https://gitlab.com/rth7680/qemu.git tags/pull-tcg-20221004 for you to fetch changes up to ab419fd8a035a65942de4e63effcd55ccbf1a9fe: target/sh4: Fix TB_FLAG_UNALIGN (2022-10-04 12:33:05 -0700) ---------------------------------------------------------------- Cache CPUClass for use in hot code paths. Add CPUTLBEntryFull, probe_access_full, tlb_set_page_full. Add generic support for TARGET_TB_PCREL. tcg/ppc: Optimize 26-bit jumps using STQ for POWER 2.07 target/sh4: Fix TB_FLAG_UNALIGN ---------------------------------------------------------------- Alex Bennée (3): cpu: cache CPUClass in CPUState for hot code paths hw/core/cpu-sysemu: used cached class in cpu_asidx_from_attrs cputlb: used cached CPUClass in our hot-paths Leandro Lupori (1): tcg/ppc: Optimize 26-bit jumps Richard Henderson (16): accel/tcg: Rename CPUIOTLBEntry to CPUTLBEntryFull accel/tcg: Drop addr member from SavedIOTLB accel/tcg: Suppress auto-invalidate in probe_access_internal accel/tcg: Introduce probe_access_full accel/tcg: Introduce tlb_set_page_full include/exec: Introduce TARGET_PAGE_ENTRY_EXTRA accel/tcg: Remove PageDesc code_bitmap accel/tcg: Use bool for page_find_alloc accel/tcg: Use DisasContextBase in plugin_gen_tb_start accel/tcg: Do not align tb->page_addr[0] accel/tcg: Inline tb_flush_jmp_cache include/hw/core: Create struct CPUJumpCache hw/core: Add CPUClass.get_pc accel/tcg: Introduce tb_pc and log_pc accel/tcg: Introduce TARGET_TB_PCREL target/sh4: Fix TB_FLAG_UNALIGN accel/tcg/internal.h | 10 ++ accel/tcg/tb-hash.h | 1 + accel/tcg/tb-jmp-cache.h | 65 ++++++++ include/exec/cpu-common.h | 1 + include/exec/cpu-defs.h | 48 ++++-- include/exec/exec-all.h | 75 ++++++++- include/exec/plugin-gen.h | 7 +- include/hw/core/cpu.h | 28 ++-- include/qemu/typedefs.h | 2 + include/tcg/tcg.h | 2 +- target/sh4/cpu.h | 56 ++++--- accel/stubs/tcg-stub.c | 4 + accel/tcg/cpu-exec.c | 80 +++++----- accel/tcg/cputlb.c | 259 ++++++++++++++++++-------------- accel/tcg/plugin-gen.c | 22 +-- accel/tcg/translate-all.c | 214 ++++++++++++-------------- accel/tcg/translator.c | 2 +- cpu.c | 9 +- hw/core/cpu-common.c | 3 +- hw/core/cpu-sysemu.c | 5 +- linux-user/sh4/signal.c | 6 +- plugins/core.c | 2 +- target/alpha/cpu.c | 9 ++ target/arm/cpu.c | 17 ++- target/arm/mte_helper.c | 14 +- target/arm/sve_helper.c | 4 +- target/arm/translate-a64.c | 2 +- target/avr/cpu.c | 10 +- target/cris/cpu.c | 8 + target/hexagon/cpu.c | 10 +- target/hppa/cpu.c | 12 +- target/i386/cpu.c | 9 ++ target/i386/tcg/tcg-cpu.c | 2 +- target/loongarch/cpu.c | 11 +- target/m68k/cpu.c | 8 + target/microblaze/cpu.c | 10 +- target/mips/cpu.c | 8 + target/mips/tcg/exception.c | 2 +- target/mips/tcg/sysemu/special_helper.c | 2 +- target/nios2/cpu.c | 9 ++ target/openrisc/cpu.c | 10 +- target/ppc/cpu_init.c | 8 + target/riscv/cpu.c | 17 ++- target/rx/cpu.c | 10 +- target/s390x/cpu.c | 8 + target/s390x/tcg/mem_helper.c | 4 - target/sh4/cpu.c | 18 ++- target/sh4/helper.c | 6 +- target/sh4/translate.c | 90 +++++------ target/sparc/cpu.c | 10 +- target/tricore/cpu.c | 11 +- target/xtensa/cpu.c | 8 + tcg/tcg.c | 8 +- trace/control-target.c | 2 +- tcg/ppc/tcg-target.c.inc | 119 +++++++++++---- 55 files changed, 915 insertions(+), 462 deletions(-) create mode 100644 accel/tcg/tb-jmp-cache.h From: Alex Bennée <alex.bennee@linaro.org> The class cast checkers are quite expensive and always on (unlike the dynamic case who's checks are gated by CONFIG_QOM_CAST_DEBUG). To avoid the overhead of repeatedly checking something which should never change we cache the CPUClass reference for use in the hot code paths. Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-Id: <20220811151413.3350684-3-alex.bennee@linaro.org> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-Id: <20220923084803.498337-3-clg@kaod.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/hw/core/cpu.h | 9 +++++++++ cpu.c | 9 ++++----- 2 files changed, 13 insertions(+), 5 deletions(-) diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/core/cpu.h +++ b/include/hw/core/cpu.h @@ -XXX,XX +XXX,XX @@ typedef int (*WriteCoreDumpFunction)(const void *buf, size_t size, */ #define CPU(obj) ((CPUState *)(obj)) +/* + * The class checkers bring in CPU_GET_CLASS() which is potentially + * expensive given the eventual call to + * object_class_dynamic_cast_assert(). Because of this the CPUState + * has a cached value for the class in cs->cc which is set up in + * cpu_exec_realizefn() for use in hot code paths. + */ typedef struct CPUClass CPUClass; DECLARE_CLASS_CHECKERS(CPUClass, CPU, TYPE_CPU) @@ -XXX,XX +XXX,XX @@ struct qemu_work_item; struct CPUState { /*< private >*/ DeviceState parent_obj; + /* cache to avoid expensive CPU_GET_CLASS */ + CPUClass *cc; /*< public >*/ int nr_cores; diff --git a/cpu.c b/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/cpu.c +++ b/cpu.c @@ -XXX,XX +XXX,XX @@ const VMStateDescription vmstate_cpu_common = { void cpu_exec_realizefn(CPUState *cpu, Error **errp) { -#ifndef CONFIG_USER_ONLY - CPUClass *cc = CPU_GET_CLASS(cpu); -#endif + /* cache the cpu class for the hotpath */ + cpu->cc = CPU_GET_CLASS(cpu); cpu_list_add(cpu); if (!accel_cpu_realizefn(cpu, errp)) { @@ -XXX,XX +XXX,XX @@ void cpu_exec_realizefn(CPUState *cpu, Error **errp) if (qdev_get_vmsd(DEVICE(cpu)) == NULL) { vmstate_register(NULL, cpu->cpu_index, &vmstate_cpu_common, cpu); } - if (cc->sysemu_ops->legacy_vmsd != NULL) { - vmstate_register(NULL, cpu->cpu_index, cc->sysemu_ops->legacy_vmsd, cpu); + if (cpu->cc->sysemu_ops->legacy_vmsd != NULL) { + vmstate_register(NULL, cpu->cpu_index, cpu->cc->sysemu_ops->legacy_vmsd, cpu); } #endif /* CONFIG_USER_ONLY */ } -- 2.34.1 From: Alex Bennée <alex.bennee@linaro.org> This is a heavily used function so lets avoid the cost of CPU_GET_CLASS. On the romulus-bmc run it has a modest effect: Before: 36.812 s ± 0.506 s After: 35.912 s ± 0.168 s Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-Id: <20220811151413.3350684-4-alex.bennee@linaro.org> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-Id: <20220923084803.498337-4-clg@kaod.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- hw/core/cpu-sysemu.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/hw/core/cpu-sysemu.c b/hw/core/cpu-sysemu.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/cpu-sysemu.c +++ b/hw/core/cpu-sysemu.c @@ -XXX,XX +XXX,XX @@ hwaddr cpu_get_phys_page_debug(CPUState *cpu, vaddr addr) int cpu_asidx_from_attrs(CPUState *cpu, MemTxAttrs attrs) { - CPUClass *cc = CPU_GET_CLASS(cpu); int ret = 0; - if (cc->sysemu_ops->asidx_from_attrs) { - ret = cc->sysemu_ops->asidx_from_attrs(cpu, attrs); + if (cpu->cc->sysemu_ops->asidx_from_attrs) { + ret = cpu->cc->sysemu_ops->asidx_from_attrs(cpu, attrs); assert(ret < cpu->num_ases && ret >= 0); } return ret; -- 2.34.1 From: Alex Bennée <alex.bennee@linaro.org> Before: 35.912 s ± 0.168 s After: 35.565 s ± 0.087 s Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-Id: <20220811151413.3350684-5-alex.bennee@linaro.org> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-Id: <20220923084803.498337-5-clg@kaod.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/cputlb.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ void tlb_set_page(CPUState *cpu, target_ulong vaddr, static void tlb_fill(CPUState *cpu, target_ulong addr, int size, MMUAccessType access_type, int mmu_idx, uintptr_t retaddr) { - CPUClass *cc = CPU_GET_CLASS(cpu); bool ok; /* * This is not a probe, so only valid return is success; failure * should result in exception + longjmp to the cpu loop. */ - ok = cc->tcg_ops->tlb_fill(cpu, addr, size, - access_type, mmu_idx, false, retaddr); + ok = cpu->cc->tcg_ops->tlb_fill(cpu, addr, size, + access_type, mmu_idx, false, retaddr); assert(ok); } @@ -XXX,XX +XXX,XX @@ static inline void cpu_unaligned_access(CPUState *cpu, vaddr addr, MMUAccessType access_type, int mmu_idx, uintptr_t retaddr) { - CPUClass *cc = CPU_GET_CLASS(cpu); - - cc->tcg_ops->do_unaligned_access(cpu, addr, access_type, mmu_idx, retaddr); + cpu->cc->tcg_ops->do_unaligned_access(cpu, addr, access_type, + mmu_idx, retaddr); } static inline void cpu_transaction_failed(CPUState *cpu, hwaddr physaddr, @@ -XXX,XX +XXX,XX @@ static int probe_access_internal(CPUArchState *env, target_ulong addr, if (!tlb_hit_page(tlb_addr, page_addr)) { if (!victim_tlb_hit(env, mmu_idx, index, elt_ofs, page_addr)) { CPUState *cs = env_cpu(env); - CPUClass *cc = CPU_GET_CLASS(cs); - if (!cc->tcg_ops->tlb_fill(cs, addr, fault_size, access_type, - mmu_idx, nonfault, retaddr)) { + if (!cs->cc->tcg_ops->tlb_fill(cs, addr, fault_size, access_type, + mmu_idx, nonfault, retaddr)) { /* Non-faulting page table read failed. */ *phost = NULL; return TLB_INVALID_MASK; -- 2.34.1 This structure will shortly contain more than just data for accessing MMIO. Rename the 'addr' member to 'xlat_section' to more clearly indicate its purpose. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/cpu-defs.h | 22 ++++---- accel/tcg/cputlb.c | 102 +++++++++++++++++++------------------ target/arm/mte_helper.c | 14 ++--- target/arm/sve_helper.c | 4 +- target/arm/translate-a64.c | 2 +- 5 files changed, 73 insertions(+), 71 deletions(-) diff --git a/include/exec/cpu-defs.h b/include/exec/cpu-defs.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/cpu-defs.h +++ b/include/exec/cpu-defs.h @@ -XXX,XX +XXX,XX @@ typedef uint64_t target_ulong; # endif # endif +/* Minimalized TLB entry for use by TCG fast path. */ typedef struct CPUTLBEntry { /* bit TARGET_LONG_BITS to TARGET_PAGE_BITS : virtual address bit TARGET_PAGE_BITS-1..4 : Nonzero for accesses that should not @@ -XXX,XX +XXX,XX @@ typedef struct CPUTLBEntry { QEMU_BUILD_BUG_ON(sizeof(CPUTLBEntry) != (1 << CPU_TLB_ENTRY_BITS)); -/* The IOTLB is not accessed directly inline by generated TCG code, - * so the CPUIOTLBEntry layout is not as critical as that of the - * CPUTLBEntry. (This is also why we don't want to combine the two - * structs into one.) +/* + * The full TLB entry, which is not accessed by generated TCG code, + * so the layout is not as critical as that of CPUTLBEntry. This is + * also why we don't want to combine the two structs. */ -typedef struct CPUIOTLBEntry { +typedef struct CPUTLBEntryFull { /* - * @addr contains: + * @xlat_section contains: * - in the lower TARGET_PAGE_BITS, a physical section number * - with the lower TARGET_PAGE_BITS masked off, an offset which * must be added to the virtual address to obtain: @@ -XXX,XX +XXX,XX @@ typedef struct CPUIOTLBEntry { * number is PHYS_SECTION_NOTDIRTY or PHYS_SECTION_ROM) * + the offset within the target MemoryRegion (otherwise) */ - hwaddr addr; + hwaddr xlat_section; MemTxAttrs attrs; -} CPUIOTLBEntry; +} CPUTLBEntryFull; /* * Data elements that are per MMU mode, minus the bits accessed by @@ -XXX,XX +XXX,XX @@ typedef struct CPUTLBDesc { size_t vindex; /* The tlb victim table, in two parts. */ CPUTLBEntry vtable[CPU_VTLB_SIZE]; - CPUIOTLBEntry viotlb[CPU_VTLB_SIZE]; - /* The iotlb. */ - CPUIOTLBEntry *iotlb; + CPUTLBEntryFull vfulltlb[CPU_VTLB_SIZE]; + CPUTLBEntryFull *fulltlb; } CPUTLBDesc; /* diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ static void tlb_mmu_resize_locked(CPUTLBDesc *desc, CPUTLBDescFast *fast, } g_free(fast->table); - g_free(desc->iotlb); + g_free(desc->fulltlb); tlb_window_reset(desc, now, 0); /* desc->n_used_entries is cleared by the caller */ fast->mask = (new_size - 1) << CPU_TLB_ENTRY_BITS; fast->table = g_try_new(CPUTLBEntry, new_size); - desc->iotlb = g_try_new(CPUIOTLBEntry, new_size); + desc->fulltlb = g_try_new(CPUTLBEntryFull, new_size); /* * If the allocations fail, try smaller sizes. We just freed some @@ -XXX,XX +XXX,XX @@ static void tlb_mmu_resize_locked(CPUTLBDesc *desc, CPUTLBDescFast *fast, * allocations to fail though, so we progressively reduce the allocation * size, aborting if we cannot even allocate the smallest TLB we support. */ - while (fast->table == NULL || desc->iotlb == NULL) { + while (fast->table == NULL || desc->fulltlb == NULL) { if (new_size == (1 << CPU_TLB_DYN_MIN_BITS)) { error_report("%s: %s", __func__, strerror(errno)); abort(); @@ -XXX,XX +XXX,XX @@ static void tlb_mmu_resize_locked(CPUTLBDesc *desc, CPUTLBDescFast *fast, fast->mask = (new_size - 1) << CPU_TLB_ENTRY_BITS; g_free(fast->table); - g_free(desc->iotlb); + g_free(desc->fulltlb); fast->table = g_try_new(CPUTLBEntry, new_size); - desc->iotlb = g_try_new(CPUIOTLBEntry, new_size); + desc->fulltlb = g_try_new(CPUTLBEntryFull, new_size); } } @@ -XXX,XX +XXX,XX @@ static void tlb_mmu_init(CPUTLBDesc *desc, CPUTLBDescFast *fast, int64_t now) desc->n_used_entries = 0; fast->mask = (n_entries - 1) << CPU_TLB_ENTRY_BITS; fast->table = g_new(CPUTLBEntry, n_entries); - desc->iotlb = g_new(CPUIOTLBEntry, n_entries); + desc->fulltlb = g_new(CPUTLBEntryFull, n_entries); tlb_mmu_flush_locked(desc, fast); } @@ -XXX,XX +XXX,XX @@ void tlb_destroy(CPUState *cpu) CPUTLBDescFast *fast = &env_tlb(env)->f[i]; g_free(fast->table); - g_free(desc->iotlb); + g_free(desc->fulltlb); } } @@ -XXX,XX +XXX,XX @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, /* Evict the old entry into the victim tlb. */ copy_tlb_helper_locked(tv, te); - desc->viotlb[vidx] = desc->iotlb[index]; + desc->vfulltlb[vidx] = desc->fulltlb[index]; tlb_n_used_entries_dec(env, mmu_idx); } @@ -XXX,XX +XXX,XX @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, * subtract here is that of the page base, and not the same as the * vaddr we add back in io_readx()/io_writex()/get_page_addr_code(). */ - desc->iotlb[index].addr = iotlb - vaddr_page; - desc->iotlb[index].attrs = attrs; + desc->fulltlb[index].xlat_section = iotlb - vaddr_page; + desc->fulltlb[index].attrs = attrs; /* Now calculate the new entry */ tn.addend = addend - vaddr_page; @@ -XXX,XX +XXX,XX @@ static inline void cpu_transaction_failed(CPUState *cpu, hwaddr physaddr, } } -static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry, +static uint64_t io_readx(CPUArchState *env, CPUTLBEntryFull *full, int mmu_idx, target_ulong addr, uintptr_t retaddr, MMUAccessType access_type, MemOp op) { @@ -XXX,XX +XXX,XX @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry, bool locked = false; MemTxResult r; - section = iotlb_to_section(cpu, iotlbentry->addr, iotlbentry->attrs); + section = iotlb_to_section(cpu, full->xlat_section, full->attrs); mr = section->mr; - mr_offset = (iotlbentry->addr & TARGET_PAGE_MASK) + addr; + mr_offset = (full->xlat_section & TARGET_PAGE_MASK) + addr; cpu->mem_io_pc = retaddr; if (!cpu->can_do_io) { cpu_io_recompile(cpu, retaddr); @@ -XXX,XX +XXX,XX @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry, qemu_mutex_lock_iothread(); locked = true; } - r = memory_region_dispatch_read(mr, mr_offset, &val, op, iotlbentry->attrs); + r = memory_region_dispatch_read(mr, mr_offset, &val, op, full->attrs); if (r != MEMTX_OK) { hwaddr physaddr = mr_offset + section->offset_within_address_space - section->offset_within_region; cpu_transaction_failed(cpu, physaddr, addr, memop_size(op), access_type, - mmu_idx, iotlbentry->attrs, r, retaddr); + mmu_idx, full->attrs, r, retaddr); } if (locked) { qemu_mutex_unlock_iothread(); @@ -XXX,XX +XXX,XX @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry, } /* - * Save a potentially trashed IOTLB entry for later lookup by plugin. - * This is read by tlb_plugin_lookup if the iotlb entry doesn't match + * Save a potentially trashed CPUTLBEntryFull for later lookup by plugin. + * This is read by tlb_plugin_lookup if the fulltlb entry doesn't match * because of the side effect of io_writex changing memory layout. */ static void save_iotlb_data(CPUState *cs, hwaddr addr, @@ -XXX,XX +XXX,XX @@ static void save_iotlb_data(CPUState *cs, hwaddr addr, #endif } -static void io_writex(CPUArchState *env, CPUIOTLBEntry *iotlbentry, +static void io_writex(CPUArchState *env, CPUTLBEntryFull *full, int mmu_idx, uint64_t val, target_ulong addr, uintptr_t retaddr, MemOp op) { @@ -XXX,XX +XXX,XX @@ static void io_writex(CPUArchState *env, CPUIOTLBEntry *iotlbentry, bool locked = false; MemTxResult r; - section = iotlb_to_section(cpu, iotlbentry->addr, iotlbentry->attrs); + section = iotlb_to_section(cpu, full->xlat_section, full->attrs); mr = section->mr; - mr_offset = (iotlbentry->addr & TARGET_PAGE_MASK) + addr; + mr_offset = (full->xlat_section & TARGET_PAGE_MASK) + addr; if (!cpu->can_do_io) { cpu_io_recompile(cpu, retaddr); } @@ -XXX,XX +XXX,XX @@ static void io_writex(CPUArchState *env, CPUIOTLBEntry *iotlbentry, * The memory_region_dispatch may trigger a flush/resize * so for plugins we save the iotlb_data just in case. */ - save_iotlb_data(cpu, iotlbentry->addr, section, mr_offset); + save_iotlb_data(cpu, full->xlat_section, section, mr_offset); if (!qemu_mutex_iothread_locked()) { qemu_mutex_lock_iothread(); locked = true; } - r = memory_region_dispatch_write(mr, mr_offset, val, op, iotlbentry->attrs); + r = memory_region_dispatch_write(mr, mr_offset, val, op, full->attrs); if (r != MEMTX_OK) { hwaddr physaddr = mr_offset + section->offset_within_address_space - section->offset_within_region; cpu_transaction_failed(cpu, physaddr, addr, memop_size(op), - MMU_DATA_STORE, mmu_idx, iotlbentry->attrs, r, + MMU_DATA_STORE, mmu_idx, full->attrs, r, retaddr); } if (locked) { @@ -XXX,XX +XXX,XX @@ static bool victim_tlb_hit(CPUArchState *env, size_t mmu_idx, size_t index, copy_tlb_helper_locked(vtlb, &tmptlb); qemu_spin_unlock(&env_tlb(env)->c.lock); - CPUIOTLBEntry tmpio, *io = &env_tlb(env)->d[mmu_idx].iotlb[index]; - CPUIOTLBEntry *vio = &env_tlb(env)->d[mmu_idx].viotlb[vidx]; - tmpio = *io; *io = *vio; *vio = tmpio; + CPUTLBEntryFull *f1 = &env_tlb(env)->d[mmu_idx].fulltlb[index]; + CPUTLBEntryFull *f2 = &env_tlb(env)->d[mmu_idx].vfulltlb[vidx]; + CPUTLBEntryFull tmpf; + tmpf = *f1; *f1 = *f2; *f2 = tmpf; return true; } } @@ -XXX,XX +XXX,XX @@ static bool victim_tlb_hit(CPUArchState *env, size_t mmu_idx, size_t index, (ADDR) & TARGET_PAGE_MASK) static void notdirty_write(CPUState *cpu, vaddr mem_vaddr, unsigned size, - CPUIOTLBEntry *iotlbentry, uintptr_t retaddr) + CPUTLBEntryFull *full, uintptr_t retaddr) { - ram_addr_t ram_addr = mem_vaddr + iotlbentry->addr; + ram_addr_t ram_addr = mem_vaddr + full->xlat_section; trace_memory_notdirty_write_access(mem_vaddr, ram_addr, size); @@ -XXX,XX +XXX,XX @@ int probe_access_flags(CPUArchState *env, target_ulong addr, /* Handle clean RAM pages. */ if (unlikely(flags & TLB_NOTDIRTY)) { uintptr_t index = tlb_index(env, mmu_idx, addr); - CPUIOTLBEntry *iotlbentry = &env_tlb(env)->d[mmu_idx].iotlb[index]; + CPUTLBEntryFull *full = &env_tlb(env)->d[mmu_idx].fulltlb[index]; - notdirty_write(env_cpu(env), addr, 1, iotlbentry, retaddr); + notdirty_write(env_cpu(env), addr, 1, full, retaddr); flags &= ~TLB_NOTDIRTY; } @@ -XXX,XX +XXX,XX @@ void *probe_access(CPUArchState *env, target_ulong addr, int size, if (unlikely(flags & (TLB_NOTDIRTY | TLB_WATCHPOINT))) { uintptr_t index = tlb_index(env, mmu_idx, addr); - CPUIOTLBEntry *iotlbentry = &env_tlb(env)->d[mmu_idx].iotlb[index]; + CPUTLBEntryFull *full = &env_tlb(env)->d[mmu_idx].fulltlb[index]; /* Handle watchpoints. */ if (flags & TLB_WATCHPOINT) { int wp_access = (access_type == MMU_DATA_STORE ? BP_MEM_WRITE : BP_MEM_READ); cpu_check_watchpoint(env_cpu(env), addr, size, - iotlbentry->attrs, wp_access, retaddr); + full->attrs, wp_access, retaddr); } /* Handle clean RAM pages. */ if (flags & TLB_NOTDIRTY) { - notdirty_write(env_cpu(env), addr, 1, iotlbentry, retaddr); + notdirty_write(env_cpu(env), addr, 1, full, retaddr); } } @@ -XXX,XX +XXX,XX @@ tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, target_ulong addr, * should have just filled the TLB. The one corner case is io_writex * which can cause TLB flushes and potential resizing of the TLBs * losing the information we need. In those cases we need to recover - * data from a copy of the iotlbentry. As long as this always occurs + * data from a copy of the CPUTLBEntryFull. As long as this always occurs * from the same thread (which a mem callback will be) this is safe. */ @@ -XXX,XX +XXX,XX @@ bool tlb_plugin_lookup(CPUState *cpu, target_ulong addr, int mmu_idx, if (likely(tlb_hit(tlb_addr, addr))) { /* We must have an iotlb entry for MMIO */ if (tlb_addr & TLB_MMIO) { - CPUIOTLBEntry *iotlbentry; - iotlbentry = &env_tlb(env)->d[mmu_idx].iotlb[index]; + CPUTLBEntryFull *full; + full = &env_tlb(env)->d[mmu_idx].fulltlb[index]; data->is_io = true; - data->v.io.section = iotlb_to_section(cpu, iotlbentry->addr, iotlbentry->attrs); - data->v.io.offset = (iotlbentry->addr & TARGET_PAGE_MASK) + addr; + data->v.io.section = + iotlb_to_section(cpu, full->xlat_section, full->attrs); + data->v.io.offset = (full->xlat_section & TARGET_PAGE_MASK) + addr; } else { data->is_io = false; data->v.ram.hostaddr = (void *)((uintptr_t)addr + tlbe->addend); @@ -XXX,XX +XXX,XX @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr, if (unlikely(tlb_addr & TLB_NOTDIRTY)) { notdirty_write(env_cpu(env), addr, size, - &env_tlb(env)->d[mmu_idx].iotlb[index], retaddr); + &env_tlb(env)->d[mmu_idx].fulltlb[index], retaddr); } return hostaddr; @@ -XXX,XX +XXX,XX @@ load_helper(CPUArchState *env, target_ulong addr, MemOpIdx oi, /* Handle anything that isn't just a straight memory access. */ if (unlikely(tlb_addr & ~TARGET_PAGE_MASK)) { - CPUIOTLBEntry *iotlbentry; + CPUTLBEntryFull *full; bool need_swap; /* For anything that is unaligned, recurse through full_load. */ @@ -XXX,XX +XXX,XX @@ load_helper(CPUArchState *env, target_ulong addr, MemOpIdx oi, goto do_unaligned_access; } - iotlbentry = &env_tlb(env)->d[mmu_idx].iotlb[index]; + full = &env_tlb(env)->d[mmu_idx].fulltlb[index]; /* Handle watchpoints. */ if (unlikely(tlb_addr & TLB_WATCHPOINT)) { /* On watchpoint hit, this will longjmp out. */ cpu_check_watchpoint(env_cpu(env), addr, size, - iotlbentry->attrs, BP_MEM_READ, retaddr); + full->attrs, BP_MEM_READ, retaddr); } need_swap = size > 1 && (tlb_addr & TLB_BSWAP); /* Handle I/O access. */ if (likely(tlb_addr & TLB_MMIO)) { - return io_readx(env, iotlbentry, mmu_idx, addr, retaddr, + return io_readx(env, full, mmu_idx, addr, retaddr, access_type, op ^ (need_swap * MO_BSWAP)); } @@ -XXX,XX +XXX,XX @@ store_helper_unaligned(CPUArchState *env, target_ulong addr, uint64_t val, */ if (unlikely(tlb_addr & TLB_WATCHPOINT)) { cpu_check_watchpoint(env_cpu(env), addr, size - size2, - env_tlb(env)->d[mmu_idx].iotlb[index].attrs, + env_tlb(env)->d[mmu_idx].fulltlb[index].attrs, BP_MEM_WRITE, retaddr); } if (unlikely(tlb_addr2 & TLB_WATCHPOINT)) { cpu_check_watchpoint(env_cpu(env), page2, size2, - env_tlb(env)->d[mmu_idx].iotlb[index2].attrs, + env_tlb(env)->d[mmu_idx].fulltlb[index2].attrs, BP_MEM_WRITE, retaddr); } @@ -XXX,XX +XXX,XX @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val, /* Handle anything that isn't just a straight memory access. */ if (unlikely(tlb_addr & ~TARGET_PAGE_MASK)) { - CPUIOTLBEntry *iotlbentry; + CPUTLBEntryFull *full; bool need_swap; /* For anything that is unaligned, recurse through byte stores. */ @@ -XXX,XX +XXX,XX @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val, goto do_unaligned_access; } - iotlbentry = &env_tlb(env)->d[mmu_idx].iotlb[index]; + full = &env_tlb(env)->d[mmu_idx].fulltlb[index]; /* Handle watchpoints. */ if (unlikely(tlb_addr & TLB_WATCHPOINT)) { /* On watchpoint hit, this will longjmp out. */ cpu_check_watchpoint(env_cpu(env), addr, size, - iotlbentry->attrs, BP_MEM_WRITE, retaddr); + full->attrs, BP_MEM_WRITE, retaddr); } need_swap = size > 1 && (tlb_addr & TLB_BSWAP); /* Handle I/O access. */ if (tlb_addr & TLB_MMIO) { - io_writex(env, iotlbentry, mmu_idx, val, addr, retaddr, + io_writex(env, full, mmu_idx, val, addr, retaddr, op ^ (need_swap * MO_BSWAP)); return; } @@ -XXX,XX +XXX,XX @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val, /* Handle clean RAM pages. */ if (tlb_addr & TLB_NOTDIRTY) { - notdirty_write(env_cpu(env), addr, size, iotlbentry, retaddr); + notdirty_write(env_cpu(env), addr, size, full, retaddr); } haddr = (void *)((uintptr_t)addr + entry->addend); diff --git a/target/arm/mte_helper.c b/target/arm/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/mte_helper.c +++ b/target/arm/mte_helper.c @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, return tags + index; #else uintptr_t index; - CPUIOTLBEntry *iotlbentry; + CPUTLBEntryFull *full; int in_page, flags; ram_addr_t ptr_ra; hwaddr ptr_paddr, tag_paddr, xlat; @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, assert(!(flags & TLB_INVALID_MASK)); /* - * Find the iotlbentry for ptr. This *must* be present in the TLB + * Find the CPUTLBEntryFull for ptr. This *must* be present in the TLB * because we just found the mapping. * TODO: Perhaps there should be a cputlb helper that returns a * matching tlb entry + iotlb entry. @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, g_assert(tlb_hit(comparator, ptr)); } # endif - iotlbentry = &env_tlb(env)->d[ptr_mmu_idx].iotlb[index]; + full = &env_tlb(env)->d[ptr_mmu_idx].fulltlb[index]; /* If the virtual page MemAttr != Tagged, access unchecked. */ - if (!arm_tlb_mte_tagged(&iotlbentry->attrs)) { + if (!arm_tlb_mte_tagged(&full->attrs)) { return NULL; } @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, int wp = ptr_access == MMU_DATA_LOAD ? BP_MEM_READ : BP_MEM_WRITE; assert(ra != 0); cpu_check_watchpoint(env_cpu(env), ptr, ptr_size, - iotlbentry->attrs, wp, ra); + full->attrs, wp, ra); } /* @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, tag_paddr = ptr_paddr >> (LOG2_TAG_GRANULE + 1); /* Look up the address in tag space. */ - tag_asi = iotlbentry->attrs.secure ? ARMASIdx_TagS : ARMASIdx_TagNS; + tag_asi = full->attrs.secure ? ARMASIdx_TagS : ARMASIdx_TagNS; tag_as = cpu_get_address_space(env_cpu(env), tag_asi); mr = address_space_translate(tag_as, tag_paddr, &xlat, NULL, tag_access == MMU_DATA_STORE, - iotlbentry->attrs); + full->attrs); /* * Note that @mr will never be NULL. If there is nothing in the address diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ bool sve_probe_page(SVEHostPage *info, bool nofault, CPUARMState *env, g_assert(tlb_hit(comparator, addr)); # endif - CPUIOTLBEntry *iotlbentry = &env_tlb(env)->d[mmu_idx].iotlb[index]; - info->attrs = iotlbentry->attrs; + CPUTLBEntryFull *full = &env_tlb(env)->d[mmu_idx].fulltlb[index]; + info->attrs = full->attrs; } #endif diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static bool is_guarded_page(CPUARMState *env, DisasContext *s) * table entry even for that case. */ return (tlb_hit(entry->addr_code, addr) && - arm_tlb_bti_gp(&env_tlb(env)->d[mmu_idx].iotlb[index].attrs)); + arm_tlb_bti_gp(&env_tlb(env)->d[mmu_idx].fulltlb[index].attrs)); #endif } -- 2.34.1 This field is only written, not read; remove it. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/hw/core/cpu.h | 1 - accel/tcg/cputlb.c | 7 +++---- 2 files changed, 3 insertions(+), 5 deletions(-) diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/core/cpu.h +++ b/include/hw/core/cpu.h @@ -XXX,XX +XXX,XX @@ struct CPUWatchpoint { * the memory regions get moved around by io_writex. */ typedef struct SavedIOTLB { - hwaddr addr; MemoryRegionSection *section; hwaddr mr_offset; } SavedIOTLB; diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ static uint64_t io_readx(CPUArchState *env, CPUTLBEntryFull *full, * This is read by tlb_plugin_lookup if the fulltlb entry doesn't match * because of the side effect of io_writex changing memory layout. */ -static void save_iotlb_data(CPUState *cs, hwaddr addr, - MemoryRegionSection *section, hwaddr mr_offset) +static void save_iotlb_data(CPUState *cs, MemoryRegionSection *section, + hwaddr mr_offset) { #ifdef CONFIG_PLUGIN SavedIOTLB *saved = &cs->saved_iotlb; - saved->addr = addr; saved->section = section; saved->mr_offset = mr_offset; #endif @@ -XXX,XX +XXX,XX @@ static void io_writex(CPUArchState *env, CPUTLBEntryFull *full, * The memory_region_dispatch may trigger a flush/resize * so for plugins we save the iotlb_data just in case. */ - save_iotlb_data(cpu, full->xlat_section, section, mr_offset); + save_iotlb_data(cpu, section, mr_offset); if (!qemu_mutex_iothread_locked()) { qemu_mutex_lock_iothread(); -- 2.34.1 When PAGE_WRITE_INV is set when calling tlb_set_page, we immediately set TLB_INVALID_MASK in order to force tlb_fill to be called on the next lookup. Here in probe_access_internal, we have just called tlb_fill and eliminated true misses, thus the lookup must be valid. This allows us to remove a warning comment from s390x. There doesn't seem to be a reason to change the code though. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/cputlb.c | 10 +++++++++- target/s390x/tcg/mem_helper.c | 4 ---- 2 files changed, 9 insertions(+), 5 deletions(-) diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ static int probe_access_internal(CPUArchState *env, target_ulong addr, } tlb_addr = tlb_read_ofs(entry, elt_ofs); + flags = TLB_FLAGS_MASK; page_addr = addr & TARGET_PAGE_MASK; if (!tlb_hit_page(tlb_addr, page_addr)) { if (!victim_tlb_hit(env, mmu_idx, index, elt_ofs, page_addr)) { @@ -XXX,XX +XXX,XX @@ static int probe_access_internal(CPUArchState *env, target_ulong addr, /* TLB resize via tlb_fill may have moved the entry. */ entry = tlb_entry(env, mmu_idx, addr); + + /* + * With PAGE_WRITE_INV, we set TLB_INVALID_MASK immediately, + * to force the next access through tlb_fill. We've just + * called tlb_fill, so we know that this entry *is* valid. + */ + flags &= ~TLB_INVALID_MASK; } tlb_addr = tlb_read_ofs(entry, elt_ofs); } - flags = tlb_addr & TLB_FLAGS_MASK; + flags &= tlb_addr; /* Fold all "mmio-like" bits into TLB_MMIO. This is not RAM. */ if (unlikely(flags & ~(TLB_WATCHPOINT | TLB_NOTDIRTY))) { diff --git a/target/s390x/tcg/mem_helper.c b/target/s390x/tcg/mem_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/tcg/mem_helper.c +++ b/target/s390x/tcg/mem_helper.c @@ -XXX,XX +XXX,XX @@ static int s390_probe_access(CPUArchState *env, target_ulong addr, int size, #else int flags; - /* - * For !CONFIG_USER_ONLY, we cannot rely on TLB_INVALID_MASK or haddr==NULL - * to detect if there was an exception during tlb_fill(). - */ env->tlb_fill_exc = 0; flags = probe_access_flags(env, addr, access_type, mmu_idx, nonfault, phost, ra); -- 2.34.1 Add an interface to return the CPUTLBEntryFull struct that goes with the lookup. The result is not intended to be valid across multiple lookups, so the user must use the results immediately. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/exec-all.h | 15 +++++++++++++ include/qemu/typedefs.h | 1 + accel/tcg/cputlb.c | 47 +++++++++++++++++++++++++---------------- 3 files changed, 45 insertions(+), 18 deletions(-) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ int probe_access_flags(CPUArchState *env, target_ulong addr, MMUAccessType access_type, int mmu_idx, bool nonfault, void **phost, uintptr_t retaddr); +#ifndef CONFIG_USER_ONLY +/** + * probe_access_full: + * Like probe_access_flags, except also return into @pfull. + * + * The CPUTLBEntryFull structure returned via @pfull is transient + * and must be consumed or copied immediately, before any further + * access or changes to TLB @mmu_idx. + */ +int probe_access_full(CPUArchState *env, target_ulong addr, + MMUAccessType access_type, int mmu_idx, + bool nonfault, void **phost, + CPUTLBEntryFull **pfull, uintptr_t retaddr); +#endif + #define CODE_GEN_ALIGN 16 /* must be >= of the size of a icache line */ /* Estimated block size for TB allocation. */ diff --git a/include/qemu/typedefs.h b/include/qemu/typedefs.h index XXXXXXX..XXXXXXX 100644 --- a/include/qemu/typedefs.h +++ b/include/qemu/typedefs.h @@ -XXX,XX +XXX,XX @@ typedef struct ConfidentialGuestSupport ConfidentialGuestSupport; typedef struct CPUAddressSpace CPUAddressSpace; typedef struct CPUArchState CPUArchState; typedef struct CPUState CPUState; +typedef struct CPUTLBEntryFull CPUTLBEntryFull; typedef struct DeviceListener DeviceListener; typedef struct DeviceState DeviceState; typedef struct DirtyBitmapSnapshot DirtyBitmapSnapshot; diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ static void notdirty_write(CPUState *cpu, vaddr mem_vaddr, unsigned size, static int probe_access_internal(CPUArchState *env, target_ulong addr, int fault_size, MMUAccessType access_type, int mmu_idx, bool nonfault, - void **phost, uintptr_t retaddr) + void **phost, CPUTLBEntryFull **pfull, + uintptr_t retaddr) { uintptr_t index = tlb_index(env, mmu_idx, addr); CPUTLBEntry *entry = tlb_entry(env, mmu_idx, addr); @@ -XXX,XX +XXX,XX @@ static int probe_access_internal(CPUArchState *env, target_ulong addr, mmu_idx, nonfault, retaddr)) { /* Non-faulting page table read failed. */ *phost = NULL; + *pfull = NULL; return TLB_INVALID_MASK; } /* TLB resize via tlb_fill may have moved the entry. */ + index = tlb_index(env, mmu_idx, addr); entry = tlb_entry(env, mmu_idx, addr); /* @@ -XXX,XX +XXX,XX @@ static int probe_access_internal(CPUArchState *env, target_ulong addr, } flags &= tlb_addr; + *pfull = &env_tlb(env)->d[mmu_idx].fulltlb[index]; + /* Fold all "mmio-like" bits into TLB_MMIO. This is not RAM. */ if (unlikely(flags & ~(TLB_WATCHPOINT | TLB_NOTDIRTY))) { *phost = NULL; @@ -XXX,XX +XXX,XX @@ static int probe_access_internal(CPUArchState *env, target_ulong addr, return flags; } -int probe_access_flags(CPUArchState *env, target_ulong addr, - MMUAccessType access_type, int mmu_idx, - bool nonfault, void **phost, uintptr_t retaddr) +int probe_access_full(CPUArchState *env, target_ulong addr, + MMUAccessType access_type, int mmu_idx, + bool nonfault, void **phost, CPUTLBEntryFull **pfull, + uintptr_t retaddr) { - int flags; - - flags = probe_access_internal(env, addr, 0, access_type, mmu_idx, - nonfault, phost, retaddr); + int flags = probe_access_internal(env, addr, 0, access_type, mmu_idx, + nonfault, phost, pfull, retaddr); /* Handle clean RAM pages. */ if (unlikely(flags & TLB_NOTDIRTY)) { - uintptr_t index = tlb_index(env, mmu_idx, addr); - CPUTLBEntryFull *full = &env_tlb(env)->d[mmu_idx].fulltlb[index]; - - notdirty_write(env_cpu(env), addr, 1, full, retaddr); + notdirty_write(env_cpu(env), addr, 1, *pfull, retaddr); flags &= ~TLB_NOTDIRTY; } return flags; } +int probe_access_flags(CPUArchState *env, target_ulong addr, + MMUAccessType access_type, int mmu_idx, + bool nonfault, void **phost, uintptr_t retaddr) +{ + CPUTLBEntryFull *full; + + return probe_access_full(env, addr, access_type, mmu_idx, + nonfault, phost, &full, retaddr); +} + void *probe_access(CPUArchState *env, target_ulong addr, int size, MMUAccessType access_type, int mmu_idx, uintptr_t retaddr) { + CPUTLBEntryFull *full; void *host; int flags; g_assert(-(addr | TARGET_PAGE_MASK) >= size); flags = probe_access_internal(env, addr, size, access_type, mmu_idx, - false, &host, retaddr); + false, &host, &full, retaddr); /* Per the interface, size == 0 merely faults the access. */ if (size == 0) { @@ -XXX,XX +XXX,XX @@ void *probe_access(CPUArchState *env, target_ulong addr, int size, } if (unlikely(flags & (TLB_NOTDIRTY | TLB_WATCHPOINT))) { - uintptr_t index = tlb_index(env, mmu_idx, addr); - CPUTLBEntryFull *full = &env_tlb(env)->d[mmu_idx].fulltlb[index]; - /* Handle watchpoints. */ if (flags & TLB_WATCHPOINT) { int wp_access = (access_type == MMU_DATA_STORE @@ -XXX,XX +XXX,XX @@ void *probe_access(CPUArchState *env, target_ulong addr, int size, void *tlb_vaddr_to_host(CPUArchState *env, abi_ptr addr, MMUAccessType access_type, int mmu_idx) { + CPUTLBEntryFull *full; void *host; int flags; flags = probe_access_internal(env, addr, 0, access_type, - mmu_idx, true, &host, 0); + mmu_idx, true, &host, &full, 0); /* No combination of flags are expected by the caller. */ return flags ? NULL : host; @@ -XXX,XX +XXX,XX @@ void *tlb_vaddr_to_host(CPUArchState *env, abi_ptr addr, tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, target_ulong addr, void **hostp) { + CPUTLBEntryFull *full; void *p; (void)probe_access_internal(env, addr, 1, MMU_INST_FETCH, - cpu_mmu_index(env, true), false, &p, 0); + cpu_mmu_index(env, true), false, &p, &full, 0); if (p == NULL) { return -1; } -- 2.34.1 Now that we have collected all of the page data into CPUTLBEntryFull, provide an interface to record that all in one go, instead of using 4 arguments. This interface allows CPUTLBEntryFull to be extended without having to change the number of arguments. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/cpu-defs.h | 14 +++++++++++ include/exec/exec-all.h | 22 ++++++++++++++++++ accel/tcg/cputlb.c | 51 ++++++++++++++++++++++++++--------------- 3 files changed, 69 insertions(+), 18 deletions(-) diff --git a/include/exec/cpu-defs.h b/include/exec/cpu-defs.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/cpu-defs.h +++ b/include/exec/cpu-defs.h @@ -XXX,XX +XXX,XX @@ typedef struct CPUTLBEntryFull { * + the offset within the target MemoryRegion (otherwise) */ hwaddr xlat_section; + + /* + * @phys_addr contains the physical address in the address space + * given by cpu_asidx_from_attrs(cpu, @attrs). + */ + hwaddr phys_addr; + + /* @attrs contains the memory transaction attributes for the page. */ MemTxAttrs attrs; + + /* @prot contains the complete protections for the page. */ + uint8_t prot; + + /* @lg_page_size contains the log2 of the page size. */ + uint8_t lg_page_size; } CPUTLBEntryFull; /* diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ void tlb_flush_range_by_mmuidx_all_cpus_synced(CPUState *cpu, uint16_t idxmap, unsigned bits); +/** + * tlb_set_page_full: + * @cpu: CPU context + * @mmu_idx: mmu index of the tlb to modify + * @vaddr: virtual address of the entry to add + * @full: the details of the tlb entry + * + * Add an entry to @cpu tlb index @mmu_idx. All of the fields of + * @full must be filled, except for xlat_section, and constitute + * the complete description of the translated page. + * + * This is generally called by the target tlb_fill function after + * having performed a successful page table walk to find the physical + * address and attributes for the translation. + * + * At most one entry for a given virtual address is permitted. Only a + * single TARGET_PAGE_SIZE region is mapped; @full->lg_page_size is only + * used by tlb_flush_page. + */ +void tlb_set_page_full(CPUState *cpu, int mmu_idx, target_ulong vaddr, + CPUTLBEntryFull *full); + /** * tlb_set_page_with_attrs: * @cpu: CPU to add this TLB entry for diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ static void tlb_add_large_page(CPUArchState *env, int mmu_idx, env_tlb(env)->d[mmu_idx].large_page_mask = lp_mask; } -/* Add a new TLB entry. At most one entry for a given virtual address +/* + * Add a new TLB entry. At most one entry for a given virtual address * is permitted. Only a single TARGET_PAGE_SIZE region is mapped, the * supplied size is only used by tlb_flush_page. * * Called from TCG-generated code, which is under an RCU read-side * critical section. */ -void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, - hwaddr paddr, MemTxAttrs attrs, int prot, - int mmu_idx, target_ulong size) +void tlb_set_page_full(CPUState *cpu, int mmu_idx, + target_ulong vaddr, CPUTLBEntryFull *full) { CPUArchState *env = cpu->env_ptr; CPUTLB *tlb = env_tlb(env); @@ -XXX,XX +XXX,XX @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, CPUTLBEntry *te, tn; hwaddr iotlb, xlat, sz, paddr_page; target_ulong vaddr_page; - int asidx = cpu_asidx_from_attrs(cpu, attrs); - int wp_flags; + int asidx, wp_flags, prot; bool is_ram, is_romd; assert_cpu_is_self(cpu); - if (size <= TARGET_PAGE_SIZE) { + if (full->lg_page_size <= TARGET_PAGE_BITS) { sz = TARGET_PAGE_SIZE; } else { - tlb_add_large_page(env, mmu_idx, vaddr, size); - sz = size; + sz = (hwaddr)1 << full->lg_page_size; + tlb_add_large_page(env, mmu_idx, vaddr, sz); } vaddr_page = vaddr & TARGET_PAGE_MASK; - paddr_page = paddr & TARGET_PAGE_MASK; + paddr_page = full->phys_addr & TARGET_PAGE_MASK; + prot = full->prot; + asidx = cpu_asidx_from_attrs(cpu, full->attrs); section = address_space_translate_for_iotlb(cpu, asidx, paddr_page, - &xlat, &sz, attrs, &prot); + &xlat, &sz, full->attrs, &prot); assert(sz >= TARGET_PAGE_SIZE); tlb_debug("vaddr=" TARGET_FMT_lx " paddr=0x" TARGET_FMT_plx " prot=%x idx=%d\n", - vaddr, paddr, prot, mmu_idx); + vaddr, full->phys_addr, prot, mmu_idx); address = vaddr_page; - if (size < TARGET_PAGE_SIZE) { + if (full->lg_page_size < TARGET_PAGE_BITS) { /* Repeat the MMU check and TLB fill on every access. */ address |= TLB_INVALID_MASK; } - if (attrs.byte_swap) { + if (full->attrs.byte_swap) { address |= TLB_BSWAP; } @@ -XXX,XX +XXX,XX @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, * subtract here is that of the page base, and not the same as the * vaddr we add back in io_readx()/io_writex()/get_page_addr_code(). */ + desc->fulltlb[index] = *full; desc->fulltlb[index].xlat_section = iotlb - vaddr_page; - desc->fulltlb[index].attrs = attrs; + desc->fulltlb[index].phys_addr = paddr_page; + desc->fulltlb[index].prot = prot; /* Now calculate the new entry */ tn.addend = addend - vaddr_page; @@ -XXX,XX +XXX,XX @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, qemu_spin_unlock(&tlb->c.lock); } -/* Add a new TLB entry, but without specifying the memory - * transaction attributes to be used. - */ +void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, + hwaddr paddr, MemTxAttrs attrs, int prot, + int mmu_idx, target_ulong size) +{ + CPUTLBEntryFull full = { + .phys_addr = paddr, + .attrs = attrs, + .prot = prot, + .lg_page_size = ctz64(size) + }; + + assert(is_power_of_2(size)); + tlb_set_page_full(cpu, mmu_idx, vaddr, &full); +} + void tlb_set_page(CPUState *cpu, target_ulong vaddr, hwaddr paddr, int prot, int mmu_idx, target_ulong size) -- 2.34.1 Allow the target to cache items from the guest page tables. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/cpu-defs.h | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/include/exec/cpu-defs.h b/include/exec/cpu-defs.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/cpu-defs.h +++ b/include/exec/cpu-defs.h @@ -XXX,XX +XXX,XX @@ typedef struct CPUTLBEntryFull { /* @lg_page_size contains the log2 of the page size. */ uint8_t lg_page_size; + + /* + * Allow target-specific additions to this structure. + * This may be used to cache items from the guest cpu + * page tables for later use by the implementation. + */ +#ifdef TARGET_PAGE_ENTRY_EXTRA + TARGET_PAGE_ENTRY_EXTRA +#endif } CPUTLBEntryFull; /* -- 2.34.1 This bitmap is created and discarded immediately. We gain nothing by its existence. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-Id: <20220822232338.1727934-2-richard.henderson@linaro.org> --- accel/tcg/translate-all.c | 78 ++------------------------------------- 1 file changed, 4 insertions(+), 74 deletions(-) diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ #define assert_memory_lock() tcg_debug_assert(have_mmap_lock()) #endif -#define SMC_BITMAP_USE_THRESHOLD 10 - typedef struct PageDesc { /* list of TBs intersecting this ram page */ uintptr_t first_tb; -#ifdef CONFIG_SOFTMMU - /* in order to optimize self modifying code, we count the number - of lookups we do to a given page to use a bitmap */ - unsigned long *code_bitmap; - unsigned int code_write_count; -#else +#ifdef CONFIG_USER_ONLY unsigned long flags; void *target_data; #endif -#ifndef CONFIG_USER_ONLY +#ifdef CONFIG_SOFTMMU QemuSpin lock; #endif } PageDesc; @@ -XXX,XX +XXX,XX @@ void tb_htable_init(void) qht_init(&tb_ctx.htable, tb_cmp, CODE_GEN_HTABLE_SIZE, mode); } -/* call with @p->lock held */ -static inline void invalidate_page_bitmap(PageDesc *p) -{ - assert_page_locked(p); -#ifdef CONFIG_SOFTMMU - g_free(p->code_bitmap); - p->code_bitmap = NULL; - p->code_write_count = 0; -#endif -} - /* Set to NULL all the 'first_tb' fields in all PageDescs. */ static void page_flush_tb_1(int level, void **lp) { @@ -XXX,XX +XXX,XX @@ static void page_flush_tb_1(int level, void **lp) for (i = 0; i < V_L2_SIZE; ++i) { page_lock(&pd[i]); pd[i].first_tb = (uintptr_t)NULL; - invalidate_page_bitmap(pd + i); page_unlock(&pd[i]); } } else { @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) if (rm_from_page_list) { p = page_find(tb->page_addr[0] >> TARGET_PAGE_BITS); tb_page_remove(p, tb); - invalidate_page_bitmap(p); if (tb->page_addr[1] != -1) { p = page_find(tb->page_addr[1] >> TARGET_PAGE_BITS); tb_page_remove(p, tb); - invalidate_page_bitmap(p); } } @@ -XXX,XX +XXX,XX @@ void tb_phys_invalidate(TranslationBlock *tb, tb_page_addr_t page_addr) } } -#ifdef CONFIG_SOFTMMU -/* call with @p->lock held */ -static void build_page_bitmap(PageDesc *p) -{ - int n, tb_start, tb_end; - TranslationBlock *tb; - - assert_page_locked(p); - p->code_bitmap = bitmap_new(TARGET_PAGE_SIZE); - - PAGE_FOR_EACH_TB(p, tb, n) { - /* NOTE: this is subtle as a TB may span two physical pages */ - if (n == 0) { - /* NOTE: tb_end may be after the end of the page, but - it is not a problem */ - tb_start = tb->pc & ~TARGET_PAGE_MASK; - tb_end = tb_start + tb->size; - if (tb_end > TARGET_PAGE_SIZE) { - tb_end = TARGET_PAGE_SIZE; - } - } else { - tb_start = 0; - tb_end = ((tb->pc + tb->size) & ~TARGET_PAGE_MASK); - } - bitmap_set(p->code_bitmap, tb_start, tb_end - tb_start); - } -} -#endif - /* add the tb in the target page and protect it if necessary * * Called with mmap_lock held for user-mode emulation. @@ -XXX,XX +XXX,XX @@ static inline void tb_page_add(PageDesc *p, TranslationBlock *tb, page_already_protected = p->first_tb != (uintptr_t)NULL; #endif p->first_tb = (uintptr_t)tb | n; - invalidate_page_bitmap(p); #if defined(CONFIG_USER_ONLY) /* translator_loop() must have made all TB pages non-writable */ @@ -XXX,XX +XXX,XX @@ tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, /* remove TB from the page(s) if we couldn't insert it */ if (unlikely(existing_tb)) { tb_page_remove(p, tb); - invalidate_page_bitmap(p); if (p2) { tb_page_remove(p2, tb); - invalidate_page_bitmap(p2); } tb = existing_tb; } @@ -XXX,XX +XXX,XX @@ tb_invalidate_phys_page_range__locked(struct page_collection *pages, #if !defined(CONFIG_USER_ONLY) /* if no code remaining, no need to continue to use slow writes */ if (!p->first_tb) { - invalidate_page_bitmap(p); tlb_unprotect_code(start); } #endif @@ -XXX,XX +XXX,XX @@ void tb_invalidate_phys_page_fast(struct page_collection *pages, } assert_page_locked(p); - if (!p->code_bitmap && - ++p->code_write_count >= SMC_BITMAP_USE_THRESHOLD) { - build_page_bitmap(p); - } - if (p->code_bitmap) { - unsigned int nr; - unsigned long b; - - nr = start & ~TARGET_PAGE_MASK; - b = p->code_bitmap[BIT_WORD(nr)] >> (nr & (BITS_PER_LONG - 1)); - if (b & ((1 << len) - 1)) { - goto do_invalidate; - } - } else { - do_invalidate: - tb_invalidate_phys_page_range__locked(pages, p, start, start + len, - retaddr); - } + tb_invalidate_phys_page_range__locked(pages, p, start, start + len, + retaddr); } #else /* Called with mmap_lock held. If pc is not 0 then it indicates the -- 2.34.1 Bool is more appropriate type for the alloc parameter. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/translate-all.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ void page_init(void) #endif } -static PageDesc *page_find_alloc(tb_page_addr_t index, int alloc) +static PageDesc *page_find_alloc(tb_page_addr_t index, bool alloc) { PageDesc *pd; void **lp; @@ -XXX,XX +XXX,XX @@ static PageDesc *page_find_alloc(tb_page_addr_t index, int alloc) static inline PageDesc *page_find(tb_page_addr_t index) { - return page_find_alloc(index, 0); + return page_find_alloc(index, false); } static void page_lock_pair(PageDesc **ret_p1, tb_page_addr_t phys1, - PageDesc **ret_p2, tb_page_addr_t phys2, int alloc); + PageDesc **ret_p2, tb_page_addr_t phys2, bool alloc); /* In user-mode page locks aren't used; mmap_lock is enough */ #ifdef CONFIG_USER_ONLY @@ -XXX,XX +XXX,XX @@ static inline void page_unlock(PageDesc *pd) /* lock the page(s) of a TB in the correct acquisition order */ static inline void page_lock_tb(const TranslationBlock *tb) { - page_lock_pair(NULL, tb->page_addr[0], NULL, tb->page_addr[1], 0); + page_lock_pair(NULL, tb->page_addr[0], NULL, tb->page_addr[1], false); } static inline void page_unlock_tb(const TranslationBlock *tb) @@ -XXX,XX +XXX,XX @@ void page_collection_unlock(struct page_collection *set) #endif /* !CONFIG_USER_ONLY */ static void page_lock_pair(PageDesc **ret_p1, tb_page_addr_t phys1, - PageDesc **ret_p2, tb_page_addr_t phys2, int alloc) + PageDesc **ret_p2, tb_page_addr_t phys2, bool alloc) { PageDesc *p1, *p2; tb_page_addr_t page1; @@ -XXX,XX +XXX,XX @@ tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, * Note that inserting into the hash table first isn't an option, since * we can only insert TBs that are fully initialized. */ - page_lock_pair(&p, phys_pc, &p2, phys_page2, 1); + page_lock_pair(&p, phys_pc, &p2, phys_page2, true); tb_page_add(p, tb, 0, phys_pc & TARGET_PAGE_MASK); if (p2) { tb_page_add(p2, tb, 1, phys_page2); @@ -XXX,XX +XXX,XX @@ void page_set_flags(target_ulong start, target_ulong end, int flags) for (addr = start, len = end - start; len != 0; len -= TARGET_PAGE_SIZE, addr += TARGET_PAGE_SIZE) { - PageDesc *p = page_find_alloc(addr >> TARGET_PAGE_BITS, 1); + PageDesc *p = page_find_alloc(addr >> TARGET_PAGE_BITS, true); /* If the write protection bit is set, then we invalidate the code inside. */ -- 2.34.1 Use the pc coming from db->pc_first rather than the TB. Use the cached host_addr rather than re-computing for the first page. We still need a separate lookup for the second page because it won't be computed for DisasContextBase until the translator actually performs a read from the page. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/plugin-gen.h | 7 ++++--- accel/tcg/plugin-gen.c | 22 +++++++++++----------- accel/tcg/translator.c | 2 +- 3 files changed, 16 insertions(+), 15 deletions(-) diff --git a/include/exec/plugin-gen.h b/include/exec/plugin-gen.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/plugin-gen.h +++ b/include/exec/plugin-gen.h @@ -XXX,XX +XXX,XX @@ struct DisasContextBase; #ifdef CONFIG_PLUGIN -bool plugin_gen_tb_start(CPUState *cpu, const TranslationBlock *tb, bool supress); +bool plugin_gen_tb_start(CPUState *cpu, const struct DisasContextBase *db, + bool supress); void plugin_gen_tb_end(CPUState *cpu); void plugin_gen_insn_start(CPUState *cpu, const struct DisasContextBase *db); void plugin_gen_insn_end(void); @@ -XXX,XX +XXX,XX @@ static inline void plugin_insn_append(abi_ptr pc, const void *from, size_t size) #else /* !CONFIG_PLUGIN */ -static inline -bool plugin_gen_tb_start(CPUState *cpu, const TranslationBlock *tb, bool supress) +static inline bool +plugin_gen_tb_start(CPUState *cpu, const struct DisasContextBase *db, bool sup) { return false; } diff --git a/accel/tcg/plugin-gen.c b/accel/tcg/plugin-gen.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/plugin-gen.c +++ b/accel/tcg/plugin-gen.c @@ -XXX,XX +XXX,XX @@ static void plugin_gen_inject(const struct qemu_plugin_tb *plugin_tb) pr_ops(); } -bool plugin_gen_tb_start(CPUState *cpu, const TranslationBlock *tb, bool mem_only) +bool plugin_gen_tb_start(CPUState *cpu, const DisasContextBase *db, + bool mem_only) { bool ret = false; @@ -XXX,XX +XXX,XX @@ bool plugin_gen_tb_start(CPUState *cpu, const TranslationBlock *tb, bool mem_onl ret = true; - ptb->vaddr = tb->pc; + ptb->vaddr = db->pc_first; ptb->vaddr2 = -1; - get_page_addr_code_hostp(cpu->env_ptr, tb->pc, &ptb->haddr1); + ptb->haddr1 = db->host_addr[0]; ptb->haddr2 = NULL; ptb->mem_only = mem_only; @@ -XXX,XX +XXX,XX @@ void plugin_gen_insn_start(CPUState *cpu, const DisasContextBase *db) * Note that we skip this when haddr1 == NULL, e.g. when we're * fetching instructions from a region not backed by RAM. */ - if (likely(ptb->haddr1 != NULL && ptb->vaddr2 == -1) && - unlikely((db->pc_next & TARGET_PAGE_MASK) != - (db->pc_first & TARGET_PAGE_MASK))) { - get_page_addr_code_hostp(cpu->env_ptr, db->pc_next, - &ptb->haddr2); - ptb->vaddr2 = db->pc_next; - } - if (likely(ptb->vaddr2 == -1)) { + if (ptb->haddr1 == NULL) { + pinsn->haddr = NULL; + } else if (is_same_page(db, db->pc_next)) { pinsn->haddr = ptb->haddr1 + pinsn->vaddr - ptb->vaddr; } else { + if (ptb->vaddr2 == -1) { + ptb->vaddr2 = TARGET_PAGE_ALIGN(db->pc_first); + get_page_addr_code_hostp(cpu->env_ptr, ptb->vaddr2, &ptb->haddr2); + } pinsn->haddr = ptb->haddr2 + pinsn->vaddr - ptb->vaddr2; } } diff --git a/accel/tcg/translator.c b/accel/tcg/translator.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translator.c +++ b/accel/tcg/translator.c @@ -XXX,XX +XXX,XX @@ void translator_loop(CPUState *cpu, TranslationBlock *tb, int max_insns, ops->tb_start(db, cpu); tcg_debug_assert(db->is_jmp == DISAS_NEXT); /* no early exit */ - plugin_enabled = plugin_gen_tb_start(cpu, tb, cflags & CF_MEMI_ONLY); + plugin_enabled = plugin_gen_tb_start(cpu, db, cflags & CF_MEMI_ONLY); while (true) { db->num_insns++; -- 2.34.1 Let tb->page_addr[0] contain the address of the first byte of the translated block, rather than the address of the page containing the start of the translated block. We need to recover this value anyway at various points, and it is easier to discard a page offset when it is not needed, which happens naturally via the existing find_page shift. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/cpu-exec.c | 16 ++++++++-------- accel/tcg/cputlb.c | 3 ++- accel/tcg/translate-all.c | 9 +++++---- 3 files changed, 15 insertions(+), 13 deletions(-) diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ struct tb_desc { target_ulong pc; target_ulong cs_base; CPUArchState *env; - tb_page_addr_t phys_page1; + tb_page_addr_t page_addr0; uint32_t flags; uint32_t cflags; uint32_t trace_vcpu_dstate; @@ -XXX,XX +XXX,XX @@ static bool tb_lookup_cmp(const void *p, const void *d) const struct tb_desc *desc = d; if (tb->pc == desc->pc && - tb->page_addr[0] == desc->phys_page1 && + tb->page_addr[0] == desc->page_addr0 && tb->cs_base == desc->cs_base && tb->flags == desc->flags && tb->trace_vcpu_dstate == desc->trace_vcpu_dstate && @@ -XXX,XX +XXX,XX @@ static bool tb_lookup_cmp(const void *p, const void *d) if (tb->page_addr[1] == -1) { return true; } else { - tb_page_addr_t phys_page2; - target_ulong virt_page2; + tb_page_addr_t phys_page1; + target_ulong virt_page1; /* * We know that the first page matched, and an otherwise valid TB @@ -XXX,XX +XXX,XX @@ static bool tb_lookup_cmp(const void *p, const void *d) * is different for the new TB. Therefore any exception raised * here by the faulting lookup is not premature. */ - virt_page2 = TARGET_PAGE_ALIGN(desc->pc); - phys_page2 = get_page_addr_code(desc->env, virt_page2); - if (tb->page_addr[1] == phys_page2) { + virt_page1 = TARGET_PAGE_ALIGN(desc->pc); + phys_page1 = get_page_addr_code(desc->env, virt_page1); + if (tb->page_addr[1] == phys_page1) { return true; } } @@ -XXX,XX +XXX,XX @@ static TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, if (phys_pc == -1) { return NULL; } - desc.phys_page1 = phys_pc & TARGET_PAGE_MASK; + desc.page_addr0 = phys_pc; h = tb_hash_func(phys_pc, pc, flags, cflags, *cpu->trace_dstate); return qht_lookup_custom(&tb_ctx.htable, &desc, h, tb_lookup_cmp); } diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ void tlb_flush_page_bits_by_mmuidx_all_cpus_synced(CPUState *src_cpu, can be detected */ void tlb_protect_code(ram_addr_t ram_addr) { - cpu_physical_memory_test_and_clear_dirty(ram_addr, TARGET_PAGE_SIZE, + cpu_physical_memory_test_and_clear_dirty(ram_addr & TARGET_PAGE_MASK, + TARGET_PAGE_SIZE, DIRTY_MEMORY_CODE); } diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) qemu_spin_unlock(&tb->jmp_lock); /* remove the TB from the hash list */ - phys_pc = tb->page_addr[0] + (tb->pc & ~TARGET_PAGE_MASK); + phys_pc = tb->page_addr[0]; h = tb_hash_func(phys_pc, tb->pc, tb->flags, orig_cflags, tb->trace_vcpu_dstate); if (!qht_remove(&tb_ctx.htable, tb, h)) { @@ -XXX,XX +XXX,XX @@ tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, * we can only insert TBs that are fully initialized. */ page_lock_pair(&p, phys_pc, &p2, phys_page2, true); - tb_page_add(p, tb, 0, phys_pc & TARGET_PAGE_MASK); + tb_page_add(p, tb, 0, phys_pc); if (p2) { tb_page_add(p2, tb, 1, phys_page2); } else { @@ -XXX,XX +XXX,XX @@ tb_invalidate_phys_page_range__locked(struct page_collection *pages, if (n == 0) { /* NOTE: tb_end may be after the end of the page, but it is not a problem */ - tb_start = tb->page_addr[0] + (tb->pc & ~TARGET_PAGE_MASK); + tb_start = tb->page_addr[0]; tb_end = tb_start + tb->size; } else { tb_start = tb->page_addr[1]; - tb_end = tb_start + ((tb->pc + tb->size) & ~TARGET_PAGE_MASK); + tb_end = tb_start + ((tb->page_addr[0] + tb->size) + & ~TARGET_PAGE_MASK); } if (!(tb_end <= start || tb_start >= end)) { #ifdef TARGET_HAS_PRECISE_SMC -- 2.34.1 This function has two users, who use it incompatibly. In tlb_flush_page_by_mmuidx_async_0, when flushing a single page, we need to flush exactly two pages. In tlb_flush_range_by_mmuidx_async_0, when flushing a range of pages, we need to flush N+1 pages. This avoids double-flushing of jmp cache pages in a range. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/cputlb.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ static void tb_jmp_cache_clear_page(CPUState *cpu, target_ulong page_addr) } } -static void tb_flush_jmp_cache(CPUState *cpu, target_ulong addr) -{ - /* Discard jump cache entries for any tb which might potentially - overlap the flushed page. */ - tb_jmp_cache_clear_page(cpu, addr - TARGET_PAGE_SIZE); - tb_jmp_cache_clear_page(cpu, addr); -} - /** * tlb_mmu_resize_locked() - perform TLB resize bookkeeping; resize if necessary * @desc: The CPUTLBDesc portion of the TLB @@ -XXX,XX +XXX,XX @@ static void tlb_flush_page_by_mmuidx_async_0(CPUState *cpu, } qemu_spin_unlock(&env_tlb(env)->c.lock); - tb_flush_jmp_cache(cpu, addr); + /* + * Discard jump cache entries for any tb which might potentially + * overlap the flushed page, which includes the previous. + */ + tb_jmp_cache_clear_page(cpu, addr - TARGET_PAGE_SIZE); + tb_jmp_cache_clear_page(cpu, addr); } /** @@ -XXX,XX +XXX,XX @@ static void tlb_flush_range_by_mmuidx_async_0(CPUState *cpu, return; } - for (target_ulong i = 0; i < d.len; i += TARGET_PAGE_SIZE) { - tb_flush_jmp_cache(cpu, d.addr + i); + /* + * Discard jump cache entries for any tb which might potentially + * overlap the flushed pages, which includes the previous. + */ + d.addr -= TARGET_PAGE_SIZE; + for (target_ulong i = 0, n = d.len / TARGET_PAGE_SIZE + 1; i < n; i++) { + tb_jmp_cache_clear_page(cpu, d.addr); + d.addr += TARGET_PAGE_SIZE; } } -- 2.34.1 Wrap the bare TranslationBlock pointer into a structure. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/tb-hash.h | 1 + accel/tcg/tb-jmp-cache.h | 24 ++++++++++++++++++++++++ include/exec/cpu-common.h | 1 + include/hw/core/cpu.h | 15 +-------------- include/qemu/typedefs.h | 1 + accel/stubs/tcg-stub.c | 4 ++++ accel/tcg/cpu-exec.c | 10 +++++++--- accel/tcg/cputlb.c | 9 +++++---- accel/tcg/translate-all.c | 28 +++++++++++++++++++++++++--- hw/core/cpu-common.c | 3 +-- plugins/core.c | 2 +- trace/control-target.c | 2 +- 12 files changed, 72 insertions(+), 28 deletions(-) create mode 100644 accel/tcg/tb-jmp-cache.h diff --git a/accel/tcg/tb-hash.h b/accel/tcg/tb-hash.h index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tb-hash.h +++ b/accel/tcg/tb-hash.h @@ -XXX,XX +XXX,XX @@ #include "exec/cpu-defs.h" #include "exec/exec-all.h" #include "qemu/xxhash.h" +#include "tb-jmp-cache.h" #ifdef CONFIG_SOFTMMU diff --git a/accel/tcg/tb-jmp-cache.h b/accel/tcg/tb-jmp-cache.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/accel/tcg/tb-jmp-cache.h @@ -XXX,XX +XXX,XX @@ +/* + * The per-CPU TranslationBlock jump cache. + * + * Copyright (c) 2003 Fabrice Bellard + * + * SPDX-License-Identifier: GPL-2.0-or-later + */ + +#ifndef ACCEL_TCG_TB_JMP_CACHE_H +#define ACCEL_TCG_TB_JMP_CACHE_H + +#define TB_JMP_CACHE_BITS 12 +#define TB_JMP_CACHE_SIZE (1 << TB_JMP_CACHE_BITS) + +/* + * Accessed in parallel; all accesses to 'tb' must be atomic. + */ +struct CPUJumpCache { + struct { + TranslationBlock *tb; + } array[TB_JMP_CACHE_SIZE]; +}; + +#endif /* ACCEL_TCG_TB_JMP_CACHE_H */ diff --git a/include/exec/cpu-common.h b/include/exec/cpu-common.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/cpu-common.h +++ b/include/exec/cpu-common.h @@ -XXX,XX +XXX,XX @@ void cpu_list_unlock(void); unsigned int cpu_list_generation_id_get(void); void tcg_flush_softmmu_tlb(CPUState *cs); +void tcg_flush_jmp_cache(CPUState *cs); void tcg_iommu_init_notifier_list(CPUState *cpu); void tcg_iommu_free_notifier_list(CPUState *cpu); diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/core/cpu.h +++ b/include/hw/core/cpu.h @@ -XXX,XX +XXX,XX @@ struct kvm_run; struct hax_vcpu_state; struct hvf_vcpu_state; -#define TB_JMP_CACHE_BITS 12 -#define TB_JMP_CACHE_SIZE (1 << TB_JMP_CACHE_BITS) - /* work queue */ /* The union type allows passing of 64 bit target pointers on 32 bit @@ -XXX,XX +XXX,XX @@ struct CPUState { CPUArchState *env_ptr; IcountDecr *icount_decr_ptr; - /* Accessed in parallel; all accesses must be atomic */ - TranslationBlock *tb_jmp_cache[TB_JMP_CACHE_SIZE]; + CPUJumpCache *tb_jmp_cache; struct GDBRegisterState *gdb_regs; int gdb_num_regs; @@ -XXX,XX +XXX,XX @@ extern CPUTailQ cpus; extern __thread CPUState *current_cpu; -static inline void cpu_tb_jmp_cache_clear(CPUState *cpu) -{ - unsigned int i; - - for (i = 0; i < TB_JMP_CACHE_SIZE; i++) { - qatomic_set(&cpu->tb_jmp_cache[i], NULL); - } -} - /** * qemu_tcg_mttcg_enabled: * Check whether we are running MultiThread TCG or not. diff --git a/include/qemu/typedefs.h b/include/qemu/typedefs.h index XXXXXXX..XXXXXXX 100644 --- a/include/qemu/typedefs.h +++ b/include/qemu/typedefs.h @@ -XXX,XX +XXX,XX @@ typedef struct CoMutex CoMutex; typedef struct ConfidentialGuestSupport ConfidentialGuestSupport; typedef struct CPUAddressSpace CPUAddressSpace; typedef struct CPUArchState CPUArchState; +typedef struct CPUJumpCache CPUJumpCache; typedef struct CPUState CPUState; typedef struct CPUTLBEntryFull CPUTLBEntryFull; typedef struct DeviceListener DeviceListener; diff --git a/accel/stubs/tcg-stub.c b/accel/stubs/tcg-stub.c index XXXXXXX..XXXXXXX 100644 --- a/accel/stubs/tcg-stub.c +++ b/accel/stubs/tcg-stub.c @@ -XXX,XX +XXX,XX @@ void tlb_set_dirty(CPUState *cpu, target_ulong vaddr) { } +void tcg_flush_jmp_cache(CPUState *cpu) +{ +} + int probe_access_flags(CPUArchState *env, target_ulong addr, MMUAccessType access_type, int mmu_idx, bool nonfault, void **phost, uintptr_t retaddr) diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ #include "sysemu/replay.h" #include "sysemu/tcg.h" #include "exec/helper-proto.h" +#include "tb-jmp-cache.h" #include "tb-hash.h" #include "tb-context.h" #include "internal.h" @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, tcg_debug_assert(!(cflags & CF_INVALID)); hash = tb_jmp_cache_hash_func(pc); - tb = qatomic_rcu_read(&cpu->tb_jmp_cache[hash]); + tb = qatomic_rcu_read(&cpu->tb_jmp_cache->array[hash].tb); if (likely(tb && tb->pc == pc && @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, if (tb == NULL) { return NULL; } - qatomic_set(&cpu->tb_jmp_cache[hash], tb); + qatomic_set(&cpu->tb_jmp_cache->array[hash].tb, tb); return tb; } @@ -XXX,XX +XXX,XX @@ int cpu_exec(CPUState *cpu) tb = tb_lookup(cpu, pc, cs_base, flags, cflags); if (tb == NULL) { + uint32_t h; + mmap_lock(); tb = tb_gen_code(cpu, pc, cs_base, flags, cflags); mmap_unlock(); @@ -XXX,XX +XXX,XX @@ int cpu_exec(CPUState *cpu) * We add the TB in the virtual pc hash table * for the fast lookup */ - qatomic_set(&cpu->tb_jmp_cache[tb_jmp_cache_hash_func(pc)], tb); + h = tb_jmp_cache_hash_func(pc); + qatomic_set(&cpu->tb_jmp_cache->array[h].tb, tb); } #ifndef CONFIG_USER_ONLY diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ static void tlb_window_reset(CPUTLBDesc *desc, int64_t ns, static void tb_jmp_cache_clear_page(CPUState *cpu, target_ulong page_addr) { - unsigned int i, i0 = tb_jmp_cache_hash_page(page_addr); + int i, i0 = tb_jmp_cache_hash_page(page_addr); + CPUJumpCache *jc = cpu->tb_jmp_cache; for (i = 0; i < TB_JMP_PAGE_SIZE; i++) { - qatomic_set(&cpu->tb_jmp_cache[i0 + i], NULL); + qatomic_set(&jc->array[i0 + i].tb, NULL); } } @@ -XXX,XX +XXX,XX @@ static void tlb_flush_by_mmuidx_async_work(CPUState *cpu, run_on_cpu_data data) qemu_spin_unlock(&env_tlb(env)->c.lock); - cpu_tb_jmp_cache_clear(cpu); + tcg_flush_jmp_cache(cpu); if (to_clean == ALL_MMUIDX_BITS) { qatomic_set(&env_tlb(env)->c.full_flush_count, @@ -XXX,XX +XXX,XX @@ static void tlb_flush_range_by_mmuidx_async_0(CPUState *cpu, * longer to clear each entry individually than it will to clear it all. */ if (d.len >= (TARGET_PAGE_SIZE * TB_JMP_CACHE_SIZE)) { - cpu_tb_jmp_cache_clear(cpu); + tcg_flush_jmp_cache(cpu); return; } diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ #include "sysemu/tcg.h" #include "qapi/error.h" #include "hw/core/tcg-cpu-ops.h" +#include "tb-jmp-cache.h" #include "tb-hash.h" #include "tb-context.h" #include "internal.h" @@ -XXX,XX +XXX,XX @@ static void do_tb_flush(CPUState *cpu, run_on_cpu_data tb_flush_count) } CPU_FOREACH(cpu) { - cpu_tb_jmp_cache_clear(cpu); + tcg_flush_jmp_cache(cpu); } qht_reset_size(&tb_ctx.htable, CODE_GEN_HTABLE_SIZE); @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) /* remove the TB from the hash list */ h = tb_jmp_cache_hash_func(tb->pc); CPU_FOREACH(cpu) { - if (qatomic_read(&cpu->tb_jmp_cache[h]) == tb) { - qatomic_set(&cpu->tb_jmp_cache[h], NULL); + CPUJumpCache *jc = cpu->tb_jmp_cache; + if (qatomic_read(&jc->array[h].tb) == tb) { + qatomic_set(&jc->array[h].tb, NULL); } } @@ -XXX,XX +XXX,XX @@ int page_unprotect(target_ulong address, uintptr_t pc) } #endif /* CONFIG_USER_ONLY */ +/* + * Called by generic code at e.g. cpu reset after cpu creation, + * therefore we must be prepared to allocate the jump cache. + */ +void tcg_flush_jmp_cache(CPUState *cpu) +{ + CPUJumpCache *jc = cpu->tb_jmp_cache; + + if (likely(jc)) { + for (int i = 0; i < TB_JMP_CACHE_SIZE; i++) { + qatomic_set(&jc->array[i].tb, NULL); + } + } else { + /* This should happen once during realize, and thus never race. */ + jc = g_new0(CPUJumpCache, 1); + jc = qatomic_xchg(&cpu->tb_jmp_cache, jc); + assert(jc == NULL); + } +} + /* This is a wrapper for common code that can not use CONFIG_SOFTMMU */ void tcg_flush_softmmu_tlb(CPUState *cs) { diff --git a/hw/core/cpu-common.c b/hw/core/cpu-common.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/cpu-common.c +++ b/hw/core/cpu-common.c @@ -XXX,XX +XXX,XX @@ static void cpu_common_reset(DeviceState *dev) cpu->cflags_next_tb = -1; if (tcg_enabled()) { - cpu_tb_jmp_cache_clear(cpu); - + tcg_flush_jmp_cache(cpu); tcg_flush_softmmu_tlb(cpu); } } diff --git a/plugins/core.c b/plugins/core.c index XXXXXXX..XXXXXXX 100644 --- a/plugins/core.c +++ b/plugins/core.c @@ -XXX,XX +XXX,XX @@ struct qemu_plugin_ctx *plugin_id_to_ctx_locked(qemu_plugin_id_t id) static void plugin_cpu_update__async(CPUState *cpu, run_on_cpu_data data) { bitmap_copy(cpu->plugin_mask, &data.host_ulong, QEMU_PLUGIN_EV_MAX); - cpu_tb_jmp_cache_clear(cpu); + tcg_flush_jmp_cache(cpu); } static void plugin_cpu_update__locked(gpointer k, gpointer v, gpointer udata) diff --git a/trace/control-target.c b/trace/control-target.c index XXXXXXX..XXXXXXX 100644 --- a/trace/control-target.c +++ b/trace/control-target.c @@ -XXX,XX +XXX,XX @@ static void trace_event_synchronize_vcpu_state_dynamic( { bitmap_copy(vcpu->trace_dstate, vcpu->trace_dstate_delayed, CPU_TRACE_DSTATE_MAX_EVENTS); - cpu_tb_jmp_cache_clear(vcpu); + tcg_flush_jmp_cache(vcpu); } void trace_event_set_vcpu_state_dynamic(CPUState *vcpu, -- 2.34.1 Populate this new method for all targets. Always match the result that would be given by cpu_get_tb_cpu_state, as we will want these values to correspond in the logs. Reviewed-by: Taylor Simpson <tsimpson@quicinc.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> (target/sparc) Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- Cc: Eduardo Habkost <eduardo@habkost.net> (supporter:Machine core) Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com> (supporter:Machine core) Cc: "Philippe Mathieu-Daudé" <f4bug@amsat.org> (reviewer:Machine core) Cc: Yanan Wang <wangyanan55@huawei.com> (reviewer:Machine core) Cc: Michael Rolnik <mrolnik@gmail.com> (maintainer:AVR TCG CPUs) Cc: "Edgar E. Iglesias" <edgar.iglesias@gmail.com> (maintainer:CRIS TCG CPUs) Cc: Taylor Simpson <tsimpson@quicinc.com> (supporter:Hexagon TCG CPUs) Cc: Song Gao <gaosong@loongson.cn> (maintainer:LoongArch TCG CPUs) Cc: Xiaojuan Yang <yangxiaojuan@loongson.cn> (maintainer:LoongArch TCG CPUs) Cc: Laurent Vivier <laurent@vivier.eu> (maintainer:M68K TCG CPUs) Cc: Jiaxun Yang <jiaxun.yang@flygoat.com> (reviewer:MIPS TCG CPUs) Cc: Aleksandar Rikalo <aleksandar.rikalo@syrmia.com> (reviewer:MIPS TCG CPUs) Cc: Chris Wulff <crwulff@gmail.com> (maintainer:NiosII TCG CPUs) Cc: Marek Vasut <marex@denx.de> (maintainer:NiosII TCG CPUs) Cc: Stafford Horne <shorne@gmail.com> (odd fixer:OpenRISC TCG CPUs) Cc: Yoshinori Sato <ysato@users.sourceforge.jp> (reviewer:RENESAS RX CPUs) Cc: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> (maintainer:SPARC TCG CPUs) Cc: Bastian Koppelmann <kbastian@mail.uni-paderborn.de> (maintainer:TriCore TCG CPUs) Cc: Max Filippov <jcmvbkbc@gmail.com> (maintainer:Xtensa TCG CPUs) Cc: qemu-arm@nongnu.org (open list:ARM TCG CPUs) Cc: qemu-ppc@nongnu.org (open list:PowerPC TCG CPUs) Cc: qemu-riscv@nongnu.org (open list:RISC-V TCG CPUs) Cc: qemu-s390x@nongnu.org (open list:S390 TCG CPUs) --- include/hw/core/cpu.h | 3 +++ target/alpha/cpu.c | 9 +++++++++ target/arm/cpu.c | 13 +++++++++++++ target/avr/cpu.c | 8 ++++++++ target/cris/cpu.c | 8 ++++++++ target/hexagon/cpu.c | 8 ++++++++ target/hppa/cpu.c | 8 ++++++++ target/i386/cpu.c | 9 +++++++++ target/loongarch/cpu.c | 9 +++++++++ target/m68k/cpu.c | 8 ++++++++ target/microblaze/cpu.c | 8 ++++++++ target/mips/cpu.c | 8 ++++++++ target/nios2/cpu.c | 9 +++++++++ target/openrisc/cpu.c | 8 ++++++++ target/ppc/cpu_init.c | 8 ++++++++ target/riscv/cpu.c | 13 +++++++++++++ target/rx/cpu.c | 8 ++++++++ target/s390x/cpu.c | 8 ++++++++ target/sh4/cpu.c | 8 ++++++++ target/sparc/cpu.c | 8 ++++++++ target/tricore/cpu.c | 9 +++++++++ target/xtensa/cpu.c | 8 ++++++++ 22 files changed, 186 insertions(+) diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/core/cpu.h +++ b/include/hw/core/cpu.h @@ -XXX,XX +XXX,XX @@ struct SysemuCPUOps; * If the target behaviour here is anything other than "set * the PC register to the value passed in" then the target must * also implement the synchronize_from_tb hook. + * @get_pc: Callback for getting the Program Counter register. + * As above, with the semantics of the target architecture. * @gdb_read_register: Callback for letting GDB read a register. * @gdb_write_register: Callback for letting GDB write a register. * @gdb_adjust_breakpoint: Callback for adjusting the address of a @@ -XXX,XX +XXX,XX @@ struct CPUClass { void (*dump_state)(CPUState *cpu, FILE *, int flags); int64_t (*get_arch_id)(CPUState *cpu); void (*set_pc)(CPUState *cpu, vaddr value); + vaddr (*get_pc)(CPUState *cpu); int (*gdb_read_register)(CPUState *cpu, GByteArray *buf, int reg); int (*gdb_write_register)(CPUState *cpu, uint8_t *buf, int reg); vaddr (*gdb_adjust_breakpoint)(CPUState *cpu, vaddr addr); diff --git a/target/alpha/cpu.c b/target/alpha/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/alpha/cpu.c +++ b/target/alpha/cpu.c @@ -XXX,XX +XXX,XX @@ static void alpha_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.pc = value; } +static vaddr alpha_cpu_get_pc(CPUState *cs) +{ + AlphaCPU *cpu = ALPHA_CPU(cs); + + return cpu->env.pc; +} + + static bool alpha_cpu_has_work(CPUState *cs) { /* Here we are checking to see if the CPU should wake up from HALT. @@ -XXX,XX +XXX,XX @@ static void alpha_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = alpha_cpu_has_work; cc->dump_state = alpha_cpu_dump_state; cc->set_pc = alpha_cpu_set_pc; + cc->get_pc = alpha_cpu_get_pc; cc->gdb_read_register = alpha_cpu_gdb_read_register; cc->gdb_write_register = alpha_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ static void arm_cpu_set_pc(CPUState *cs, vaddr value) } } +static vaddr arm_cpu_get_pc(CPUState *cs) +{ + ARMCPU *cpu = ARM_CPU(cs); + CPUARMState *env = &cpu->env; + + if (is_a64(env)) { + return env->pc; + } else { + return env->regs[15]; + } +} + #ifdef CONFIG_TCG void arm_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) @@ -XXX,XX +XXX,XX @@ static void arm_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = arm_cpu_has_work; cc->dump_state = arm_cpu_dump_state; cc->set_pc = arm_cpu_set_pc; + cc->get_pc = arm_cpu_get_pc; cc->gdb_read_register = arm_cpu_gdb_read_register; cc->gdb_write_register = arm_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/avr/cpu.c b/target/avr/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/avr/cpu.c +++ b/target/avr/cpu.c @@ -XXX,XX +XXX,XX @@ static void avr_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.pc_w = value / 2; /* internally PC points to words */ } +static vaddr avr_cpu_get_pc(CPUState *cs) +{ + AVRCPU *cpu = AVR_CPU(cs); + + return cpu->env.pc_w * 2; +} + static bool avr_cpu_has_work(CPUState *cs) { AVRCPU *cpu = AVR_CPU(cs); @@ -XXX,XX +XXX,XX @@ static void avr_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = avr_cpu_has_work; cc->dump_state = avr_cpu_dump_state; cc->set_pc = avr_cpu_set_pc; + cc->get_pc = avr_cpu_get_pc; dc->vmsd = &vms_avr_cpu; cc->sysemu_ops = &avr_sysemu_ops; cc->disas_set_info = avr_cpu_disas_set_info; diff --git a/target/cris/cpu.c b/target/cris/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/cris/cpu.c +++ b/target/cris/cpu.c @@ -XXX,XX +XXX,XX @@ static void cris_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.pc = value; } +static vaddr cris_cpu_get_pc(CPUState *cs) +{ + CRISCPU *cpu = CRIS_CPU(cs); + + return cpu->env.pc; +} + static bool cris_cpu_has_work(CPUState *cs) { return cs->interrupt_request & (CPU_INTERRUPT_HARD | CPU_INTERRUPT_NMI); @@ -XXX,XX +XXX,XX @@ static void cris_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = cris_cpu_has_work; cc->dump_state = cris_cpu_dump_state; cc->set_pc = cris_cpu_set_pc; + cc->get_pc = cris_cpu_get_pc; cc->gdb_read_register = cris_cpu_gdb_read_register; cc->gdb_write_register = cris_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/hexagon/cpu.c b/target/hexagon/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/cpu.c +++ b/target/hexagon/cpu.c @@ -XXX,XX +XXX,XX @@ static void hexagon_cpu_set_pc(CPUState *cs, vaddr value) env->gpr[HEX_REG_PC] = value; } +static vaddr hexagon_cpu_get_pc(CPUState *cs) +{ + HexagonCPU *cpu = HEXAGON_CPU(cs); + CPUHexagonState *env = &cpu->env; + return env->gpr[HEX_REG_PC]; +} + static void hexagon_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static void hexagon_cpu_class_init(ObjectClass *c, void *data) cc->has_work = hexagon_cpu_has_work; cc->dump_state = hexagon_dump_state; cc->set_pc = hexagon_cpu_set_pc; + cc->get_pc = hexagon_cpu_get_pc; cc->gdb_read_register = hexagon_gdb_read_register; cc->gdb_write_register = hexagon_gdb_write_register; cc->gdb_num_core_regs = TOTAL_PER_THREAD_REGS + NUM_VREGS + NUM_QREGS; diff --git a/target/hppa/cpu.c b/target/hppa/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hppa/cpu.c +++ b/target/hppa/cpu.c @@ -XXX,XX +XXX,XX @@ static void hppa_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.iaoq_b = value + 4; } +static vaddr hppa_cpu_get_pc(CPUState *cs) +{ + HPPACPU *cpu = HPPA_CPU(cs); + + return cpu->env.iaoq_f; +} + static void hppa_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static void hppa_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = hppa_cpu_has_work; cc->dump_state = hppa_cpu_dump_state; cc->set_pc = hppa_cpu_set_pc; + cc->get_pc = hppa_cpu_get_pc; cc->gdb_read_register = hppa_cpu_gdb_read_register; cc->gdb_write_register = hppa_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/i386/cpu.c b/target/i386/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -XXX,XX +XXX,XX @@ static void x86_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.eip = value; } +static vaddr x86_cpu_get_pc(CPUState *cs) +{ + X86CPU *cpu = X86_CPU(cs); + + /* Match cpu_get_tb_cpu_state. */ + return cpu->env.eip + cpu->env.segs[R_CS].base; +} + int x86_cpu_pending_interrupt(CPUState *cs, int interrupt_request) { X86CPU *cpu = X86_CPU(cs); @@ -XXX,XX +XXX,XX @@ static void x86_cpu_common_class_init(ObjectClass *oc, void *data) cc->has_work = x86_cpu_has_work; cc->dump_state = x86_cpu_dump_state; cc->set_pc = x86_cpu_set_pc; + cc->get_pc = x86_cpu_get_pc; cc->gdb_read_register = x86_cpu_gdb_read_register; cc->gdb_write_register = x86_cpu_gdb_write_register; cc->get_arch_id = x86_cpu_get_arch_id; diff --git a/target/loongarch/cpu.c b/target/loongarch/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/loongarch/cpu.c +++ b/target/loongarch/cpu.c @@ -XXX,XX +XXX,XX @@ static void loongarch_cpu_set_pc(CPUState *cs, vaddr value) env->pc = value; } +static vaddr loongarch_cpu_get_pc(CPUState *cs) +{ + LoongArchCPU *cpu = LOONGARCH_CPU(cs); + CPULoongArchState *env = &cpu->env; + + return env->pc; +} + #ifndef CONFIG_USER_ONLY #include "hw/loongarch/virt.h" @@ -XXX,XX +XXX,XX @@ static void loongarch_cpu_class_init(ObjectClass *c, void *data) cc->has_work = loongarch_cpu_has_work; cc->dump_state = loongarch_cpu_dump_state; cc->set_pc = loongarch_cpu_set_pc; + cc->get_pc = loongarch_cpu_get_pc; #ifndef CONFIG_USER_ONLY dc->vmsd = &vmstate_loongarch_cpu; cc->sysemu_ops = &loongarch_sysemu_ops; diff --git a/target/m68k/cpu.c b/target/m68k/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/m68k/cpu.c +++ b/target/m68k/cpu.c @@ -XXX,XX +XXX,XX @@ static void m68k_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.pc = value; } +static vaddr m68k_cpu_get_pc(CPUState *cs) +{ + M68kCPU *cpu = M68K_CPU(cs); + + return cpu->env.pc; +} + static bool m68k_cpu_has_work(CPUState *cs) { return cs->interrupt_request & CPU_INTERRUPT_HARD; @@ -XXX,XX +XXX,XX @@ static void m68k_cpu_class_init(ObjectClass *c, void *data) cc->has_work = m68k_cpu_has_work; cc->dump_state = m68k_cpu_dump_state; cc->set_pc = m68k_cpu_set_pc; + cc->get_pc = m68k_cpu_get_pc; cc->gdb_read_register = m68k_cpu_gdb_read_register; cc->gdb_write_register = m68k_cpu_gdb_write_register; #if defined(CONFIG_SOFTMMU) diff --git a/target/microblaze/cpu.c b/target/microblaze/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/microblaze/cpu.c +++ b/target/microblaze/cpu.c @@ -XXX,XX +XXX,XX @@ static void mb_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.iflags = 0; } +static vaddr mb_cpu_get_pc(CPUState *cs) +{ + MicroBlazeCPU *cpu = MICROBLAZE_CPU(cs); + + return cpu->env.pc; +} + static void mb_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static void mb_cpu_class_init(ObjectClass *oc, void *data) cc->dump_state = mb_cpu_dump_state; cc->set_pc = mb_cpu_set_pc; + cc->get_pc = mb_cpu_get_pc; cc->gdb_read_register = mb_cpu_gdb_read_register; cc->gdb_write_register = mb_cpu_gdb_write_register; diff --git a/target/mips/cpu.c b/target/mips/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/cpu.c +++ b/target/mips/cpu.c @@ -XXX,XX +XXX,XX @@ static void mips_cpu_set_pc(CPUState *cs, vaddr value) mips_env_set_pc(&cpu->env, value); } +static vaddr mips_cpu_get_pc(CPUState *cs) +{ + MIPSCPU *cpu = MIPS_CPU(cs); + + return cpu->env.active_tc.PC; +} + static bool mips_cpu_has_work(CPUState *cs) { MIPSCPU *cpu = MIPS_CPU(cs); @@ -XXX,XX +XXX,XX @@ static void mips_cpu_class_init(ObjectClass *c, void *data) cc->has_work = mips_cpu_has_work; cc->dump_state = mips_cpu_dump_state; cc->set_pc = mips_cpu_set_pc; + cc->get_pc = mips_cpu_get_pc; cc->gdb_read_register = mips_cpu_gdb_read_register; cc->gdb_write_register = mips_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/nios2/cpu.c b/target/nios2/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/nios2/cpu.c +++ b/target/nios2/cpu.c @@ -XXX,XX +XXX,XX @@ static void nios2_cpu_set_pc(CPUState *cs, vaddr value) env->pc = value; } +static vaddr nios2_cpu_get_pc(CPUState *cs) +{ + Nios2CPU *cpu = NIOS2_CPU(cs); + CPUNios2State *env = &cpu->env; + + return env->pc; +} + static bool nios2_cpu_has_work(CPUState *cs) { return cs->interrupt_request & CPU_INTERRUPT_HARD; @@ -XXX,XX +XXX,XX @@ static void nios2_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = nios2_cpu_has_work; cc->dump_state = nios2_cpu_dump_state; cc->set_pc = nios2_cpu_set_pc; + cc->get_pc = nios2_cpu_get_pc; cc->disas_set_info = nios2_cpu_disas_set_info; #ifndef CONFIG_USER_ONLY cc->sysemu_ops = &nios2_sysemu_ops; diff --git a/target/openrisc/cpu.c b/target/openrisc/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/openrisc/cpu.c +++ b/target/openrisc/cpu.c @@ -XXX,XX +XXX,XX @@ static void openrisc_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.dflag = 0; } +static vaddr openrisc_cpu_get_pc(CPUState *cs) +{ + OpenRISCCPU *cpu = OPENRISC_CPU(cs); + + return cpu->env.pc; +} + static void openrisc_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static void openrisc_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = openrisc_cpu_has_work; cc->dump_state = openrisc_cpu_dump_state; cc->set_pc = openrisc_cpu_set_pc; + cc->get_pc = openrisc_cpu_get_pc; cc->gdb_read_register = openrisc_cpu_gdb_read_register; cc->gdb_write_register = openrisc_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/ppc/cpu_init.c b/target/ppc/cpu_init.c index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/cpu_init.c +++ b/target/ppc/cpu_init.c @@ -XXX,XX +XXX,XX @@ static void ppc_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.nip = value; } +static vaddr ppc_cpu_get_pc(CPUState *cs) +{ + PowerPCCPU *cpu = POWERPC_CPU(cs); + + return cpu->env.nip; +} + static bool ppc_cpu_has_work(CPUState *cs) { PowerPCCPU *cpu = POWERPC_CPU(cs); @@ -XXX,XX +XXX,XX @@ static void ppc_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = ppc_cpu_has_work; cc->dump_state = ppc_cpu_dump_state; cc->set_pc = ppc_cpu_set_pc; + cc->get_pc = ppc_cpu_get_pc; cc->gdb_read_register = ppc_cpu_gdb_read_register; cc->gdb_write_register = ppc_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_set_pc(CPUState *cs, vaddr value) } } +static vaddr riscv_cpu_get_pc(CPUState *cs) +{ + RISCVCPU *cpu = RISCV_CPU(cs); + CPURISCVState *env = &cpu->env; + + /* Match cpu_get_tb_cpu_state. */ + if (env->xl == MXL_RV32) { + return env->pc & UINT32_MAX; + } + return env->pc; +} + static void riscv_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_class_init(ObjectClass *c, void *data) cc->has_work = riscv_cpu_has_work; cc->dump_state = riscv_cpu_dump_state; cc->set_pc = riscv_cpu_set_pc; + cc->get_pc = riscv_cpu_get_pc; cc->gdb_read_register = riscv_cpu_gdb_read_register; cc->gdb_write_register = riscv_cpu_gdb_write_register; cc->gdb_num_core_regs = 33; diff --git a/target/rx/cpu.c b/target/rx/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/rx/cpu.c +++ b/target/rx/cpu.c @@ -XXX,XX +XXX,XX @@ static void rx_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.pc = value; } +static vaddr rx_cpu_get_pc(CPUState *cs) +{ + RXCPU *cpu = RX_CPU(cs); + + return cpu->env.pc; +} + static void rx_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static void rx_cpu_class_init(ObjectClass *klass, void *data) cc->has_work = rx_cpu_has_work; cc->dump_state = rx_cpu_dump_state; cc->set_pc = rx_cpu_set_pc; + cc->get_pc = rx_cpu_get_pc; #ifndef CONFIG_USER_ONLY cc->sysemu_ops = &rx_sysemu_ops; diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/cpu.c +++ b/target/s390x/cpu.c @@ -XXX,XX +XXX,XX @@ static void s390_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.psw.addr = value; } +static vaddr s390_cpu_get_pc(CPUState *cs) +{ + S390CPU *cpu = S390_CPU(cs); + + return cpu->env.psw.addr; +} + static bool s390_cpu_has_work(CPUState *cs) { S390CPU *cpu = S390_CPU(cs); @@ -XXX,XX +XXX,XX @@ static void s390_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = s390_cpu_has_work; cc->dump_state = s390_cpu_dump_state; cc->set_pc = s390_cpu_set_pc; + cc->get_pc = s390_cpu_get_pc; cc->gdb_read_register = s390_cpu_gdb_read_register; cc->gdb_write_register = s390_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/sh4/cpu.c b/target/sh4/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/cpu.c +++ b/target/sh4/cpu.c @@ -XXX,XX +XXX,XX @@ static void superh_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.pc = value; } +static vaddr superh_cpu_get_pc(CPUState *cs) +{ + SuperHCPU *cpu = SUPERH_CPU(cs); + + return cpu->env.pc; +} + static void superh_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static void superh_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = superh_cpu_has_work; cc->dump_state = superh_cpu_dump_state; cc->set_pc = superh_cpu_set_pc; + cc->get_pc = superh_cpu_get_pc; cc->gdb_read_register = superh_cpu_gdb_read_register; cc->gdb_write_register = superh_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/sparc/cpu.c b/target/sparc/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/cpu.c +++ b/target/sparc/cpu.c @@ -XXX,XX +XXX,XX @@ static void sparc_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.npc = value + 4; } +static vaddr sparc_cpu_get_pc(CPUState *cs) +{ + SPARCCPU *cpu = SPARC_CPU(cs); + + return cpu->env.pc; +} + static void sparc_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static void sparc_cpu_class_init(ObjectClass *oc, void *data) cc->memory_rw_debug = sparc_cpu_memory_rw_debug; #endif cc->set_pc = sparc_cpu_set_pc; + cc->get_pc = sparc_cpu_get_pc; cc->gdb_read_register = sparc_cpu_gdb_read_register; cc->gdb_write_register = sparc_cpu_gdb_write_register; #ifndef CONFIG_USER_ONLY diff --git a/target/tricore/cpu.c b/target/tricore/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/tricore/cpu.c +++ b/target/tricore/cpu.c @@ -XXX,XX +XXX,XX @@ static void tricore_cpu_set_pc(CPUState *cs, vaddr value) env->PC = value & ~(target_ulong)1; } +static vaddr tricore_cpu_get_pc(CPUState *cs) +{ + TriCoreCPU *cpu = TRICORE_CPU(cs); + CPUTriCoreState *env = &cpu->env; + + return env->PC; +} + static void tricore_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { @@ -XXX,XX +XXX,XX @@ static void tricore_cpu_class_init(ObjectClass *c, void *data) cc->dump_state = tricore_cpu_dump_state; cc->set_pc = tricore_cpu_set_pc; + cc->get_pc = tricore_cpu_get_pc; cc->sysemu_ops = &tricore_sysemu_ops; cc->tcg_ops = &tricore_tcg_ops; } diff --git a/target/xtensa/cpu.c b/target/xtensa/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/xtensa/cpu.c +++ b/target/xtensa/cpu.c @@ -XXX,XX +XXX,XX @@ static void xtensa_cpu_set_pc(CPUState *cs, vaddr value) cpu->env.pc = value; } +static vaddr xtensa_cpu_get_pc(CPUState *cs) +{ + XtensaCPU *cpu = XTENSA_CPU(cs); + + return cpu->env.pc; +} + static bool xtensa_cpu_has_work(CPUState *cs) { #ifndef CONFIG_USER_ONLY @@ -XXX,XX +XXX,XX @@ static void xtensa_cpu_class_init(ObjectClass *oc, void *data) cc->has_work = xtensa_cpu_has_work; cc->dump_state = xtensa_cpu_dump_state; cc->set_pc = xtensa_cpu_set_pc; + cc->get_pc = xtensa_cpu_get_pc; cc->gdb_read_register = xtensa_cpu_gdb_read_register; cc->gdb_write_register = xtensa_cpu_gdb_write_register; cc->gdb_stop_before_watchpoint = true; -- 2.34.1 The availability of tb->pc will shortly be conditional. Introduce accessor functions to minimize ifdefs. Pass around a known pc to places like tcg_gen_code, where the caller must already have the value. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/internal.h | 6 ++++ include/exec/exec-all.h | 6 ++++ include/tcg/tcg.h | 2 +- accel/tcg/cpu-exec.c | 46 ++++++++++++++----------- accel/tcg/translate-all.c | 37 +++++++++++--------- target/arm/cpu.c | 4 +-- target/avr/cpu.c | 2 +- target/hexagon/cpu.c | 2 +- target/hppa/cpu.c | 4 +-- target/i386/tcg/tcg-cpu.c | 2 +- target/loongarch/cpu.c | 2 +- target/microblaze/cpu.c | 2 +- target/mips/tcg/exception.c | 2 +- target/mips/tcg/sysemu/special_helper.c | 2 +- target/openrisc/cpu.c | 2 +- target/riscv/cpu.c | 4 +-- target/rx/cpu.c | 2 +- target/sh4/cpu.c | 4 +-- target/sparc/cpu.c | 2 +- target/tricore/cpu.c | 2 +- tcg/tcg.c | 8 ++--- 21 files changed, 82 insertions(+), 61 deletions(-) diff --git a/accel/tcg/internal.h b/accel/tcg/internal.h index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/internal.h +++ b/accel/tcg/internal.h @@ -XXX,XX +XXX,XX @@ G_NORETURN void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr); void page_init(void); void tb_htable_init(void); +/* Return the current PC from CPU, which may be cached in TB. */ +static inline target_ulong log_pc(CPUState *cpu, const TranslationBlock *tb) +{ + return tb_pc(tb); +} + #endif /* ACCEL_TCG_INTERNAL_H */ diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ struct TranslationBlock { uintptr_t jmp_dest[2]; }; +/* Hide the read to avoid ifdefs for TARGET_TB_PCREL. */ +static inline target_ulong tb_pc(const TranslationBlock *tb) +{ + return tb->pc; +} + /* Hide the qatomic_read to make code a little easier on the eyes */ static inline uint32_t tb_cflags(const TranslationBlock *tb) { diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg.h +++ b/include/tcg/tcg.h @@ -XXX,XX +XXX,XX @@ void tcg_register_thread(void); void tcg_prologue_init(TCGContext *s); void tcg_func_start(TCGContext *s); -int tcg_gen_code(TCGContext *s, TranslationBlock *tb); +int tcg_gen_code(TCGContext *s, TranslationBlock *tb, target_ulong pc_start); void tcg_set_frame(TCGContext *s, TCGReg reg, intptr_t start, intptr_t size); diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ static bool tb_lookup_cmp(const void *p, const void *d) const TranslationBlock *tb = p; const struct tb_desc *desc = d; - if (tb->pc == desc->pc && + if (tb_pc(tb) == desc->pc && tb->page_addr[0] == desc->page_addr0 && tb->cs_base == desc->cs_base && tb->flags == desc->flags && @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, return tb; } -static inline void log_cpu_exec(target_ulong pc, CPUState *cpu, - const TranslationBlock *tb) +static void log_cpu_exec(target_ulong pc, CPUState *cpu, + const TranslationBlock *tb) { - if (unlikely(qemu_loglevel_mask(CPU_LOG_TB_CPU | CPU_LOG_EXEC)) - && qemu_log_in_addr_range(pc)) { - + if (qemu_log_in_addr_range(pc)) { qemu_log_mask(CPU_LOG_EXEC, "Trace %d: %p [" TARGET_FMT_lx "/" TARGET_FMT_lx "/%08x/%08x] %s\n", @@ -XXX,XX +XXX,XX @@ const void *HELPER(lookup_tb_ptr)(CPUArchState *env) return tcg_code_gen_epilogue; } - log_cpu_exec(pc, cpu, tb); + if (qemu_loglevel_mask(CPU_LOG_TB_CPU | CPU_LOG_EXEC)) { + log_cpu_exec(pc, cpu, tb); + } return tb->tc.ptr; } @@ -XXX,XX +XXX,XX @@ cpu_tb_exec(CPUState *cpu, TranslationBlock *itb, int *tb_exit) TranslationBlock *last_tb; const void *tb_ptr = itb->tc.ptr; - log_cpu_exec(itb->pc, cpu, itb); + if (qemu_loglevel_mask(CPU_LOG_TB_CPU | CPU_LOG_EXEC)) { + log_cpu_exec(log_pc(cpu, itb), cpu, itb); + } qemu_thread_jit_execute(); ret = tcg_qemu_tb_exec(env, tb_ptr); @@ -XXX,XX +XXX,XX @@ cpu_tb_exec(CPUState *cpu, TranslationBlock *itb, int *tb_exit) * of the start of the TB. */ CPUClass *cc = CPU_GET_CLASS(cpu); - qemu_log_mask_and_addr(CPU_LOG_EXEC, last_tb->pc, - "Stopped execution of TB chain before %p [" - TARGET_FMT_lx "] %s\n", - last_tb->tc.ptr, last_tb->pc, - lookup_symbol(last_tb->pc)); + if (cc->tcg_ops->synchronize_from_tb) { cc->tcg_ops->synchronize_from_tb(cpu, last_tb); } else { assert(cc->set_pc); - cc->set_pc(cpu, last_tb->pc); + cc->set_pc(cpu, tb_pc(last_tb)); + } + if (qemu_loglevel_mask(CPU_LOG_EXEC)) { + target_ulong pc = log_pc(cpu, last_tb); + if (qemu_log_in_addr_range(pc)) { + qemu_log("Stopped execution of TB chain before %p [" + TARGET_FMT_lx "] %s\n", + last_tb->tc.ptr, pc, lookup_symbol(pc)); + } } } @@ -XXX,XX +XXX,XX @@ static inline void tb_add_jump(TranslationBlock *tb, int n, qemu_spin_unlock(&tb_next->jmp_lock); - qemu_log_mask_and_addr(CPU_LOG_EXEC, tb->pc, - "Linking TBs %p [" TARGET_FMT_lx - "] index %d -> %p [" TARGET_FMT_lx "]\n", - tb->tc.ptr, tb->pc, n, - tb_next->tc.ptr, tb_next->pc); + qemu_log_mask(CPU_LOG_EXEC, "Linking TBs %p index %d -> %p\n", + tb->tc.ptr, n, tb_next->tc.ptr); return; out_unlock_next: @@ -XXX,XX +XXX,XX @@ static inline bool cpu_handle_interrupt(CPUState *cpu, } static inline void cpu_loop_exec_tb(CPUState *cpu, TranslationBlock *tb, + target_ulong pc, TranslationBlock **last_tb, int *tb_exit) { int32_t insns_left; - trace_exec_tb(tb, tb->pc); + trace_exec_tb(tb, pc); tb = cpu_tb_exec(cpu, tb, tb_exit); if (*tb_exit != TB_EXIT_REQUESTED) { *last_tb = tb; @@ -XXX,XX +XXX,XX @@ int cpu_exec(CPUState *cpu) tb_add_jump(last_tb, tb_exit, tb); } - cpu_loop_exec_tb(cpu, tb, &last_tb, &tb_exit); + cpu_loop_exec_tb(cpu, tb, pc, &last_tb, &tb_exit); /* Try to align the host and virtual clocks if the guest is in advance */ diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ static int encode_search(TranslationBlock *tb, uint8_t *block) for (j = 0; j < TARGET_INSN_START_WORDS; ++j) { if (i == 0) { - prev = (j == 0 ? tb->pc : 0); + prev = (j == 0 ? tb_pc(tb) : 0); } else { prev = tcg_ctx->gen_insn_data[i - 1][j]; } @@ -XXX,XX +XXX,XX @@ static int encode_search(TranslationBlock *tb, uint8_t *block) static int cpu_restore_state_from_tb(CPUState *cpu, TranslationBlock *tb, uintptr_t searched_pc, bool reset_icount) { - target_ulong data[TARGET_INSN_START_WORDS] = { tb->pc }; + target_ulong data[TARGET_INSN_START_WORDS] = { tb_pc(tb) }; uintptr_t host_pc = (uintptr_t)tb->tc.ptr; CPUArchState *env = cpu->env_ptr; const uint8_t *p = tb->tc.ptr + tb->tc.size; @@ -XXX,XX +XXX,XX @@ static bool tb_cmp(const void *ap, const void *bp) const TranslationBlock *a = ap; const TranslationBlock *b = bp; - return a->pc == b->pc && + return tb_pc(a) == tb_pc(b) && a->cs_base == b->cs_base && a->flags == b->flags && (tb_cflags(a) & ~CF_INVALID) == (tb_cflags(b) & ~CF_INVALID) && @@ -XXX,XX +XXX,XX @@ static void do_tb_invalidate_check(void *p, uint32_t hash, void *userp) TranslationBlock *tb = p; target_ulong addr = *(target_ulong *)userp; - if (!(addr + TARGET_PAGE_SIZE <= tb->pc || addr >= tb->pc + tb->size)) { + if (!(addr + TARGET_PAGE_SIZE <= tb_pc(tb) || + addr >= tb_pc(tb) + tb->size)) { printf("ERROR invalidate: address=" TARGET_FMT_lx - " PC=%08lx size=%04x\n", addr, (long)tb->pc, tb->size); + " PC=%08lx size=%04x\n", addr, (long)tb_pc(tb), tb->size); } } @@ -XXX,XX +XXX,XX @@ static void do_tb_page_check(void *p, uint32_t hash, void *userp) TranslationBlock *tb = p; int flags1, flags2; - flags1 = page_get_flags(tb->pc); - flags2 = page_get_flags(tb->pc + tb->size - 1); + flags1 = page_get_flags(tb_pc(tb)); + flags2 = page_get_flags(tb_pc(tb) + tb->size - 1); if ((flags1 & PAGE_WRITE) || (flags2 & PAGE_WRITE)) { printf("ERROR page flags: PC=%08lx size=%04x f1=%x f2=%x\n", - (long)tb->pc, tb->size, flags1, flags2); + (long)tb_pc(tb), tb->size, flags1, flags2); } } @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) /* remove the TB from the hash list */ phys_pc = tb->page_addr[0]; - h = tb_hash_func(phys_pc, tb->pc, tb->flags, orig_cflags, + h = tb_hash_func(phys_pc, tb_pc(tb), tb->flags, orig_cflags, tb->trace_vcpu_dstate); if (!qht_remove(&tb_ctx.htable, tb, h)) { return; @@ -XXX,XX +XXX,XX @@ tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, } /* add in the hash table */ - h = tb_hash_func(phys_pc, tb->pc, tb->flags, tb->cflags, + h = tb_hash_func(phys_pc, tb_pc(tb), tb->flags, tb->cflags, tb->trace_vcpu_dstate); qht_insert(&tb_ctx.htable, tb, h, &existing_tb); @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_gen_code(CPUState *cpu, tcg_ctx->cpu = NULL; max_insns = tb->icount; - trace_translate_block(tb, tb->pc, tb->tc.ptr); + trace_translate_block(tb, pc, tb->tc.ptr); /* generate machine code */ tb->jmp_reset_offset[0] = TB_JMP_RESET_OFFSET_INVALID; @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_gen_code(CPUState *cpu, ti = profile_getclock(); #endif - gen_code_size = tcg_gen_code(tcg_ctx, tb); + gen_code_size = tcg_gen_code(tcg_ctx, tb, pc); if (unlikely(gen_code_size < 0)) { error_return: switch (gen_code_size) { @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_gen_code(CPUState *cpu, #ifdef DEBUG_DISAS if (qemu_loglevel_mask(CPU_LOG_TB_OUT_ASM) && - qemu_log_in_addr_range(tb->pc)) { + qemu_log_in_addr_range(pc)) { FILE *logfile = qemu_log_trylock(); if (logfile) { int code_size, data_size; @@ -XXX,XX +XXX,XX @@ void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr) */ cpu->cflags_next_tb = curr_cflags(cpu) | CF_MEMI_ONLY | CF_LAST_IO | n; - qemu_log_mask_and_addr(CPU_LOG_EXEC, tb->pc, - "cpu_io_recompile: rewound execution of TB to " - TARGET_FMT_lx "\n", tb->pc); + if (qemu_loglevel_mask(CPU_LOG_EXEC)) { + target_ulong pc = log_pc(cpu, tb); + if (qemu_log_in_addr_range(pc)) { + qemu_log("cpu_io_recompile: rewound execution of TB to " + TARGET_FMT_lx "\n", pc); + } + } cpu_loop_exit_noexc(cpu); } diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ void arm_cpu_synchronize_from_tb(CPUState *cs, * never possible for an AArch64 TB to chain to an AArch32 TB. */ if (is_a64(env)) { - env->pc = tb->pc; + env->pc = tb_pc(tb); } else { - env->regs[15] = tb->pc; + env->regs[15] = tb_pc(tb); } } #endif /* CONFIG_TCG */ diff --git a/target/avr/cpu.c b/target/avr/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/avr/cpu.c +++ b/target/avr/cpu.c @@ -XXX,XX +XXX,XX @@ static void avr_cpu_synchronize_from_tb(CPUState *cs, AVRCPU *cpu = AVR_CPU(cs); CPUAVRState *env = &cpu->env; - env->pc_w = tb->pc / 2; /* internally PC points to words */ + env->pc_w = tb_pc(tb) / 2; /* internally PC points to words */ } static void avr_cpu_reset(DeviceState *ds) diff --git a/target/hexagon/cpu.c b/target/hexagon/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/cpu.c +++ b/target/hexagon/cpu.c @@ -XXX,XX +XXX,XX @@ static void hexagon_cpu_synchronize_from_tb(CPUState *cs, { HexagonCPU *cpu = HEXAGON_CPU(cs); CPUHexagonState *env = &cpu->env; - env->gpr[HEX_REG_PC] = tb->pc; + env->gpr[HEX_REG_PC] = tb_pc(tb); } static bool hexagon_cpu_has_work(CPUState *cs) diff --git a/target/hppa/cpu.c b/target/hppa/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hppa/cpu.c +++ b/target/hppa/cpu.c @@ -XXX,XX +XXX,XX @@ static void hppa_cpu_synchronize_from_tb(CPUState *cs, HPPACPU *cpu = HPPA_CPU(cs); #ifdef CONFIG_USER_ONLY - cpu->env.iaoq_f = tb->pc; + cpu->env.iaoq_f = tb_pc(tb); cpu->env.iaoq_b = tb->cs_base; #else /* Recover the IAOQ values from the GVA + PRIV. */ @@ -XXX,XX +XXX,XX @@ static void hppa_cpu_synchronize_from_tb(CPUState *cs, int32_t diff = cs_base; cpu->env.iasq_f = iasq_f; - cpu->env.iaoq_f = (tb->pc & ~iasq_f) + priv; + cpu->env.iaoq_f = (tb_pc(tb) & ~iasq_f) + priv; if (diff) { cpu->env.iaoq_b = cpu->env.iaoq_f + diff; } diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/tcg-cpu.c +++ b/target/i386/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static void x86_cpu_synchronize_from_tb(CPUState *cs, { X86CPU *cpu = X86_CPU(cs); - cpu->env.eip = tb->pc - tb->cs_base; + cpu->env.eip = tb_pc(tb) - tb->cs_base; } #ifndef CONFIG_USER_ONLY diff --git a/target/loongarch/cpu.c b/target/loongarch/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/loongarch/cpu.c +++ b/target/loongarch/cpu.c @@ -XXX,XX +XXX,XX @@ static void loongarch_cpu_synchronize_from_tb(CPUState *cs, LoongArchCPU *cpu = LOONGARCH_CPU(cs); CPULoongArchState *env = &cpu->env; - env->pc = tb->pc; + env->pc = tb_pc(tb); } #endif /* CONFIG_TCG */ diff --git a/target/microblaze/cpu.c b/target/microblaze/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/microblaze/cpu.c +++ b/target/microblaze/cpu.c @@ -XXX,XX +XXX,XX @@ static void mb_cpu_synchronize_from_tb(CPUState *cs, { MicroBlazeCPU *cpu = MICROBLAZE_CPU(cs); - cpu->env.pc = tb->pc; + cpu->env.pc = tb_pc(tb); cpu->env.iflags = tb->flags & IFLAGS_TB_MASK; } diff --git a/target/mips/tcg/exception.c b/target/mips/tcg/exception.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/exception.c +++ b/target/mips/tcg/exception.c @@ -XXX,XX +XXX,XX @@ void mips_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) MIPSCPU *cpu = MIPS_CPU(cs); CPUMIPSState *env = &cpu->env; - env->active_tc.PC = tb->pc; + env->active_tc.PC = tb_pc(tb); env->hflags &= ~MIPS_HFLAG_BMASK; env->hflags |= tb->flags & MIPS_HFLAG_BMASK; } diff --git a/target/mips/tcg/sysemu/special_helper.c b/target/mips/tcg/sysemu/special_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/sysemu/special_helper.c +++ b/target/mips/tcg/sysemu/special_helper.c @@ -XXX,XX +XXX,XX @@ bool mips_io_recompile_replay_branch(CPUState *cs, const TranslationBlock *tb) CPUMIPSState *env = &cpu->env; if ((env->hflags & MIPS_HFLAG_BMASK) != 0 - && env->active_tc.PC != tb->pc) { + && env->active_tc.PC != tb_pc(tb)) { env->active_tc.PC -= (env->hflags & MIPS_HFLAG_B16 ? 2 : 4); env->hflags &= ~MIPS_HFLAG_BMASK; return true; diff --git a/target/openrisc/cpu.c b/target/openrisc/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/openrisc/cpu.c +++ b/target/openrisc/cpu.c @@ -XXX,XX +XXX,XX @@ static void openrisc_cpu_synchronize_from_tb(CPUState *cs, { OpenRISCCPU *cpu = OPENRISC_CPU(cs); - cpu->env.pc = tb->pc; + cpu->env.pc = tb_pc(tb); } diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_synchronize_from_tb(CPUState *cs, RISCVMXL xl = FIELD_EX32(tb->flags, TB_FLAGS, XL); if (xl == MXL_RV32) { - env->pc = (int32_t)tb->pc; + env->pc = (int32_t)tb_pc(tb); } else { - env->pc = tb->pc; + env->pc = tb_pc(tb); } } diff --git a/target/rx/cpu.c b/target/rx/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/rx/cpu.c +++ b/target/rx/cpu.c @@ -XXX,XX +XXX,XX @@ static void rx_cpu_synchronize_from_tb(CPUState *cs, { RXCPU *cpu = RX_CPU(cs); - cpu->env.pc = tb->pc; + cpu->env.pc = tb_pc(tb); } static bool rx_cpu_has_work(CPUState *cs) diff --git a/target/sh4/cpu.c b/target/sh4/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/cpu.c +++ b/target/sh4/cpu.c @@ -XXX,XX +XXX,XX @@ static void superh_cpu_synchronize_from_tb(CPUState *cs, { SuperHCPU *cpu = SUPERH_CPU(cs); - cpu->env.pc = tb->pc; + cpu->env.pc = tb_pc(tb); cpu->env.flags = tb->flags & TB_FLAG_ENVFLAGS_MASK; } @@ -XXX,XX +XXX,XX @@ static bool superh_io_recompile_replay_branch(CPUState *cs, CPUSH4State *env = &cpu->env; if ((env->flags & ((DELAY_SLOT | DELAY_SLOT_CONDITIONAL))) != 0 - && env->pc != tb->pc) { + && env->pc != tb_pc(tb)) { env->pc -= 2; env->flags &= ~(DELAY_SLOT | DELAY_SLOT_CONDITIONAL); return true; diff --git a/target/sparc/cpu.c b/target/sparc/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/cpu.c +++ b/target/sparc/cpu.c @@ -XXX,XX +XXX,XX @@ static void sparc_cpu_synchronize_from_tb(CPUState *cs, { SPARCCPU *cpu = SPARC_CPU(cs); - cpu->env.pc = tb->pc; + cpu->env.pc = tb_pc(tb); cpu->env.npc = tb->cs_base; } diff --git a/target/tricore/cpu.c b/target/tricore/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/tricore/cpu.c +++ b/target/tricore/cpu.c @@ -XXX,XX +XXX,XX @@ static void tricore_cpu_synchronize_from_tb(CPUState *cs, TriCoreCPU *cpu = TRICORE_CPU(cs); CPUTriCoreState *env = &cpu->env; - env->PC = tb->pc; + env->PC = tb_pc(tb); } static void tricore_cpu_reset(DeviceState *dev) diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ int64_t tcg_cpu_exec_time(void) #endif -int tcg_gen_code(TCGContext *s, TranslationBlock *tb) +int tcg_gen_code(TCGContext *s, TranslationBlock *tb, target_ulong pc_start) { #ifdef CONFIG_PROFILER TCGProfile *prof = &s->prof; @@ -XXX,XX +XXX,XX @@ int tcg_gen_code(TCGContext *s, TranslationBlock *tb) #ifdef DEBUG_DISAS if (unlikely(qemu_loglevel_mask(CPU_LOG_TB_OP) - && qemu_log_in_addr_range(tb->pc))) { + && qemu_log_in_addr_range(pc_start))) { FILE *logfile = qemu_log_trylock(); if (logfile) { fprintf(logfile, "OP:\n"); @@ -XXX,XX +XXX,XX @@ int tcg_gen_code(TCGContext *s, TranslationBlock *tb) if (s->nb_indirects > 0) { #ifdef DEBUG_DISAS if (unlikely(qemu_loglevel_mask(CPU_LOG_TB_OP_IND) - && qemu_log_in_addr_range(tb->pc))) { + && qemu_log_in_addr_range(pc_start))) { FILE *logfile = qemu_log_trylock(); if (logfile) { fprintf(logfile, "OP before indirect lowering:\n"); @@ -XXX,XX +XXX,XX @@ int tcg_gen_code(TCGContext *s, TranslationBlock *tb) #ifdef DEBUG_DISAS if (unlikely(qemu_loglevel_mask(CPU_LOG_TB_OP_OPT) - && qemu_log_in_addr_range(tb->pc))) { + && qemu_log_in_addr_range(pc_start))) { FILE *logfile = qemu_log_trylock(); if (logfile) { fprintf(logfile, "OP after optimization and liveness analysis:\n"); -- 2.34.1 Prepare for targets to be able to produce TBs that can run in more than one virtual context. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/internal.h | 4 +++ accel/tcg/tb-jmp-cache.h | 41 +++++++++++++++++++++++++ include/exec/cpu-defs.h | 3 ++ include/exec/exec-all.h | 32 ++++++++++++++++++-- accel/tcg/cpu-exec.c | 16 ++++++---- accel/tcg/translate-all.c | 64 ++++++++++++++++++++++++++------------- 6 files changed, 131 insertions(+), 29 deletions(-) diff --git a/accel/tcg/internal.h b/accel/tcg/internal.h index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/internal.h +++ b/accel/tcg/internal.h @@ -XXX,XX +XXX,XX @@ void tb_htable_init(void); /* Return the current PC from CPU, which may be cached in TB. */ static inline target_ulong log_pc(CPUState *cpu, const TranslationBlock *tb) { +#if TARGET_TB_PCREL + return cpu->cc->get_pc(cpu); +#else return tb_pc(tb); +#endif } #endif /* ACCEL_TCG_INTERNAL_H */ diff --git a/accel/tcg/tb-jmp-cache.h b/accel/tcg/tb-jmp-cache.h index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tb-jmp-cache.h +++ b/accel/tcg/tb-jmp-cache.h @@ -XXX,XX +XXX,XX @@ /* * Accessed in parallel; all accesses to 'tb' must be atomic. + * For TARGET_TB_PCREL, accesses to 'pc' must be protected by + * a load_acquire/store_release to 'tb'. */ struct CPUJumpCache { struct { TranslationBlock *tb; +#if TARGET_TB_PCREL + target_ulong pc; +#endif } array[TB_JMP_CACHE_SIZE]; }; +static inline TranslationBlock * +tb_jmp_cache_get_tb(CPUJumpCache *jc, uint32_t hash) +{ +#if TARGET_TB_PCREL + /* Use acquire to ensure current load of pc from jc. */ + return qatomic_load_acquire(&jc->array[hash].tb); +#else + /* Use rcu_read to ensure current load of pc from *tb. */ + return qatomic_rcu_read(&jc->array[hash].tb); +#endif +} + +static inline target_ulong +tb_jmp_cache_get_pc(CPUJumpCache *jc, uint32_t hash, TranslationBlock *tb) +{ +#if TARGET_TB_PCREL + return jc->array[hash].pc; +#else + return tb_pc(tb); +#endif +} + +static inline void +tb_jmp_cache_set(CPUJumpCache *jc, uint32_t hash, + TranslationBlock *tb, target_ulong pc) +{ +#if TARGET_TB_PCREL + jc->array[hash].pc = pc; + /* Use store_release on tb to ensure pc is written first. */ + qatomic_store_release(&jc->array[hash].tb, tb); +#else + /* Use the pc value already stored in tb->pc. */ + qatomic_set(&jc->array[hash].tb, tb); +#endif +} + #endif /* ACCEL_TCG_TB_JMP_CACHE_H */ diff --git a/include/exec/cpu-defs.h b/include/exec/cpu-defs.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/cpu-defs.h +++ b/include/exec/cpu-defs.h @@ -XXX,XX +XXX,XX @@ # error TARGET_PAGE_BITS must be defined in cpu-param.h # endif #endif +#ifndef TARGET_TB_PCREL +# define TARGET_TB_PCREL 0 +#endif #define TARGET_LONG_SIZE (TARGET_LONG_BITS / 8) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ struct tb_tc { }; struct TranslationBlock { - target_ulong pc; /* simulated PC corresponding to this block (EIP + CS base) */ - target_ulong cs_base; /* CS base for this block */ +#if !TARGET_TB_PCREL + /* + * Guest PC corresponding to this block. This must be the true + * virtual address. Therefore e.g. x86 stores EIP + CS_BASE, and + * targets like Arm, MIPS, HP-PA, which reuse low bits for ISA or + * privilege, must store those bits elsewhere. + * + * If TARGET_TB_PCREL, the opcodes for the TranslationBlock are + * written such that the TB is associated only with the physical + * page and may be run in any virtual address context. In this case, + * PC must always be taken from ENV in a target-specific manner. + * Unwind information is taken as offsets from the page, to be + * deposited into the "current" PC. + */ + target_ulong pc; +#endif + + /* + * Target-specific data associated with the TranslationBlock, e.g.: + * x86: the original user, the Code Segment virtual base, + * arm: an extension of tb->flags, + * s390x: instruction data for EXECUTE, + * sparc: the next pc of the instruction queue (for delay slots). + */ + target_ulong cs_base; + uint32_t flags; /* flags defining in which context the code was generated */ uint32_t cflags; /* compile flags */ @@ -XXX,XX +XXX,XX @@ struct TranslationBlock { /* Hide the read to avoid ifdefs for TARGET_TB_PCREL. */ static inline target_ulong tb_pc(const TranslationBlock *tb) { +#if TARGET_TB_PCREL + qemu_build_not_reached(); +#else return tb->pc; +#endif } /* Hide the qatomic_read to make code a little easier on the eyes */ diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ static bool tb_lookup_cmp(const void *p, const void *d) const TranslationBlock *tb = p; const struct tb_desc *desc = d; - if (tb_pc(tb) == desc->pc && + if ((TARGET_TB_PCREL || tb_pc(tb) == desc->pc) && tb->page_addr[0] == desc->page_addr0 && tb->cs_base == desc->cs_base && tb->flags == desc->flags && @@ -XXX,XX +XXX,XX @@ static TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, return NULL; } desc.page_addr0 = phys_pc; - h = tb_hash_func(phys_pc, pc, flags, cflags, *cpu->trace_dstate); + h = tb_hash_func(phys_pc, (TARGET_TB_PCREL ? 0 : pc), + flags, cflags, *cpu->trace_dstate); return qht_lookup_custom(&tb_ctx.htable, &desc, h, tb_lookup_cmp); } @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, uint32_t flags, uint32_t cflags) { TranslationBlock *tb; + CPUJumpCache *jc; uint32_t hash; /* we should never be trying to look up an INVALID tb */ tcg_debug_assert(!(cflags & CF_INVALID)); hash = tb_jmp_cache_hash_func(pc); - tb = qatomic_rcu_read(&cpu->tb_jmp_cache->array[hash].tb); + jc = cpu->tb_jmp_cache; + tb = tb_jmp_cache_get_tb(jc, hash); if (likely(tb && - tb->pc == pc && + tb_jmp_cache_get_pc(jc, hash, tb) == pc && tb->cs_base == cs_base && tb->flags == flags && tb->trace_vcpu_dstate == *cpu->trace_dstate && @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, if (tb == NULL) { return NULL; } - qatomic_set(&cpu->tb_jmp_cache->array[hash].tb, tb); + tb_jmp_cache_set(jc, hash, tb, pc); return tb; } @@ -XXX,XX +XXX,XX @@ cpu_tb_exec(CPUState *cpu, TranslationBlock *itb, int *tb_exit) if (cc->tcg_ops->synchronize_from_tb) { cc->tcg_ops->synchronize_from_tb(cpu, last_tb); } else { + assert(!TARGET_TB_PCREL); assert(cc->set_pc); cc->set_pc(cpu, tb_pc(last_tb)); } @@ -XXX,XX +XXX,XX @@ int cpu_exec(CPUState *cpu) * for the fast lookup */ h = tb_jmp_cache_hash_func(pc); - qatomic_set(&cpu->tb_jmp_cache->array[h].tb, tb); + tb_jmp_cache_set(cpu->tb_jmp_cache, h, tb, pc); } #ifndef CONFIG_USER_ONLY diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ static int encode_search(TranslationBlock *tb, uint8_t *block) for (j = 0; j < TARGET_INSN_START_WORDS; ++j) { if (i == 0) { - prev = (j == 0 ? tb_pc(tb) : 0); + prev = (!TARGET_TB_PCREL && j == 0 ? tb_pc(tb) : 0); } else { prev = tcg_ctx->gen_insn_data[i - 1][j]; } @@ -XXX,XX +XXX,XX @@ static int encode_search(TranslationBlock *tb, uint8_t *block) static int cpu_restore_state_from_tb(CPUState *cpu, TranslationBlock *tb, uintptr_t searched_pc, bool reset_icount) { - target_ulong data[TARGET_INSN_START_WORDS] = { tb_pc(tb) }; + target_ulong data[TARGET_INSN_START_WORDS]; uintptr_t host_pc = (uintptr_t)tb->tc.ptr; CPUArchState *env = cpu->env_ptr; const uint8_t *p = tb->tc.ptr + tb->tc.size; @@ -XXX,XX +XXX,XX @@ static int cpu_restore_state_from_tb(CPUState *cpu, TranslationBlock *tb, return -1; } + memset(data, 0, sizeof(data)); + if (!TARGET_TB_PCREL) { + data[0] = tb_pc(tb); + } + /* Reconstruct the stored insn data while looking for the point at which the end of the insn exceeds the searched_pc. */ for (i = 0; i < num_insns; ++i) { @@ -XXX,XX +XXX,XX @@ static bool tb_cmp(const void *ap, const void *bp) const TranslationBlock *a = ap; const TranslationBlock *b = bp; - return tb_pc(a) == tb_pc(b) && - a->cs_base == b->cs_base && - a->flags == b->flags && - (tb_cflags(a) & ~CF_INVALID) == (tb_cflags(b) & ~CF_INVALID) && - a->trace_vcpu_dstate == b->trace_vcpu_dstate && - a->page_addr[0] == b->page_addr[0] && - a->page_addr[1] == b->page_addr[1]; + return ((TARGET_TB_PCREL || tb_pc(a) == tb_pc(b)) && + a->cs_base == b->cs_base && + a->flags == b->flags && + (tb_cflags(a) & ~CF_INVALID) == (tb_cflags(b) & ~CF_INVALID) && + a->trace_vcpu_dstate == b->trace_vcpu_dstate && + a->page_addr[0] == b->page_addr[0] && + a->page_addr[1] == b->page_addr[1]); } void tb_htable_init(void) @@ -XXX,XX +XXX,XX @@ static inline void tb_jmp_unlink(TranslationBlock *dest) qemu_spin_unlock(&dest->jmp_lock); } +static void tb_jmp_cache_inval_tb(TranslationBlock *tb) +{ + CPUState *cpu; + + if (TARGET_TB_PCREL) { + /* A TB may be at any virtual address */ + CPU_FOREACH(cpu) { + tcg_flush_jmp_cache(cpu); + } + } else { + uint32_t h = tb_jmp_cache_hash_func(tb_pc(tb)); + + CPU_FOREACH(cpu) { + CPUJumpCache *jc = cpu->tb_jmp_cache; + + if (qatomic_read(&jc->array[h].tb) == tb) { + qatomic_set(&jc->array[h].tb, NULL); + } + } + } +} + /* * In user-mode, call with mmap_lock held. * In !user-mode, if @rm_from_page_list is set, call with the TB's pages' @@ -XXX,XX +XXX,XX @@ static inline void tb_jmp_unlink(TranslationBlock *dest) */ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) { - CPUState *cpu; PageDesc *p; uint32_t h; tb_page_addr_t phys_pc; @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) /* remove the TB from the hash list */ phys_pc = tb->page_addr[0]; - h = tb_hash_func(phys_pc, tb_pc(tb), tb->flags, orig_cflags, - tb->trace_vcpu_dstate); + h = tb_hash_func(phys_pc, (TARGET_TB_PCREL ? 0 : tb_pc(tb)), + tb->flags, orig_cflags, tb->trace_vcpu_dstate); if (!qht_remove(&tb_ctx.htable, tb, h)) { return; } @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) } /* remove the TB from the hash list */ - h = tb_jmp_cache_hash_func(tb->pc); - CPU_FOREACH(cpu) { - CPUJumpCache *jc = cpu->tb_jmp_cache; - if (qatomic_read(&jc->array[h].tb) == tb) { - qatomic_set(&jc->array[h].tb, NULL); - } - } + tb_jmp_cache_inval_tb(tb); /* suppress this TB from the two jump lists */ tb_remove_from_jmp_list(tb, 0); @@ -XXX,XX +XXX,XX @@ tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, } /* add in the hash table */ - h = tb_hash_func(phys_pc, tb_pc(tb), tb->flags, tb->cflags, - tb->trace_vcpu_dstate); + h = tb_hash_func(phys_pc, (TARGET_TB_PCREL ? 0 : tb_pc(tb)), + tb->flags, tb->cflags, tb->trace_vcpu_dstate); qht_insert(&tb_ctx.htable, tb, h, &existing_tb); /* remove TB from the page(s) if we couldn't insert it */ @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_gen_code(CPUState *cpu, gen_code_buf = tcg_ctx->code_gen_ptr; tb->tc.ptr = tcg_splitwx_to_rx(gen_code_buf); +#if !TARGET_TB_PCREL tb->pc = pc; +#endif tb->cs_base = cs_base; tb->flags = flags; tb->cflags = cflags; -- 2.34.1 From: Leandro Lupori <leandro.lupori@eldorado.org.br> PowerPC64 processors handle direct branches better than indirect ones, resulting in less stalled cycles and branch misses. However, PPC's tb_target_set_jmp_target() was only using direct branches for 16-bit jumps, while PowerPC64's unconditional branch instructions are able to handle displacements of up to 26 bits. To take advantage of this, now jumps whose displacements fit in between 17 and 26 bits are also converted to direct branches. Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Leandro Lupori <leandro.lupori@eldorado.org.br> [rth: Expanded some commentary.] Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/ppc/tcg-target.c.inc | 119 +++++++++++++++++++++++++++++---------- 1 file changed, 88 insertions(+), 31 deletions(-) diff --git a/tcg/ppc/tcg-target.c.inc b/tcg/ppc/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/ppc/tcg-target.c.inc +++ b/tcg/ppc/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void tcg_out_mb(TCGContext *s, TCGArg a0) tcg_out32(s, insn); } +static inline uint64_t make_pair(tcg_insn_unit i1, tcg_insn_unit i2) +{ + if (HOST_BIG_ENDIAN) { + return (uint64_t)i1 << 32 | i2; + } + return (uint64_t)i2 << 32 | i1; +} + +static inline void ppc64_replace2(uintptr_t rx, uintptr_t rw, + tcg_insn_unit i0, tcg_insn_unit i1) +{ +#if TCG_TARGET_REG_BITS == 64 + qatomic_set((uint64_t *)rw, make_pair(i0, i1)); + flush_idcache_range(rx, rw, 8); +#else + qemu_build_not_reached(); +#endif +} + +static inline void ppc64_replace4(uintptr_t rx, uintptr_t rw, + tcg_insn_unit i0, tcg_insn_unit i1, + tcg_insn_unit i2, tcg_insn_unit i3) +{ + uint64_t p[2]; + + p[!HOST_BIG_ENDIAN] = make_pair(i0, i1); + p[HOST_BIG_ENDIAN] = make_pair(i2, i3); + + /* + * There's no convenient way to get the compiler to allocate a pair + * of registers at an even index, so copy into r6/r7 and clobber. + */ + asm("mr %%r6, %1\n\t" + "mr %%r7, %2\n\t" + "stq %%r6, %0" + : "=Q"(*(__int128 *)rw) : "r"(p[0]), "r"(p[1]) : "r6", "r7"); + flush_idcache_range(rx, rw, 16); +} + void tb_target_set_jmp_target(uintptr_t tc_ptr, uintptr_t jmp_rx, uintptr_t jmp_rw, uintptr_t addr) { - if (TCG_TARGET_REG_BITS == 64) { - tcg_insn_unit i1, i2; - intptr_t tb_diff = addr - tc_ptr; - intptr_t br_diff = addr - (jmp_rx + 4); - uint64_t pair; + tcg_insn_unit i0, i1, i2, i3; + intptr_t tb_diff = addr - tc_ptr; + intptr_t br_diff = addr - (jmp_rx + 4); + intptr_t lo, hi; - /* This does not exercise the range of the branch, but we do - still need to be able to load the new value of TCG_REG_TB. - But this does still happen quite often. */ - if (tb_diff == (int16_t)tb_diff) { - i1 = ADDI | TAI(TCG_REG_TB, TCG_REG_TB, tb_diff); - i2 = B | (br_diff & 0x3fffffc); - } else { - intptr_t lo = (int16_t)tb_diff; - intptr_t hi = (int32_t)(tb_diff - lo); - assert(tb_diff == hi + lo); - i1 = ADDIS | TAI(TCG_REG_TB, TCG_REG_TB, hi >> 16); - i2 = ADDI | TAI(TCG_REG_TB, TCG_REG_TB, lo); - } -#if HOST_BIG_ENDIAN - pair = (uint64_t)i1 << 32 | i2; -#else - pair = (uint64_t)i2 << 32 | i1; -#endif - - /* As per the enclosing if, this is ppc64. Avoid the _Static_assert - within qatomic_set that would fail to build a ppc32 host. */ - qatomic_set__nocheck((uint64_t *)jmp_rw, pair); - flush_idcache_range(jmp_rx, jmp_rw, 8); - } else { + if (TCG_TARGET_REG_BITS == 32) { intptr_t diff = addr - jmp_rx; tcg_debug_assert(in_range_b(diff)); qatomic_set((uint32_t *)jmp_rw, B | (diff & 0x3fffffc)); flush_idcache_range(jmp_rx, jmp_rw, 4); + return; } + + /* + * For 16-bit displacements, we can use a single add + branch. + * This happens quite often. + */ + if (tb_diff == (int16_t)tb_diff) { + i0 = ADDI | TAI(TCG_REG_TB, TCG_REG_TB, tb_diff); + i1 = B | (br_diff & 0x3fffffc); + ppc64_replace2(jmp_rx, jmp_rw, i0, i1); + return; + } + + lo = (int16_t)tb_diff; + hi = (int32_t)(tb_diff - lo); + assert(tb_diff == hi + lo); + i0 = ADDIS | TAI(TCG_REG_TB, TCG_REG_TB, hi >> 16); + i1 = ADDI | TAI(TCG_REG_TB, TCG_REG_TB, lo); + + /* + * Without stq from 2.07, we can only update two insns, + * and those must be the ones that load the target address. + */ + if (!have_isa_2_07) { + ppc64_replace2(jmp_rx, jmp_rw, i0, i1); + return; + } + + /* + * For 26-bit displacements, we can use a direct branch. + * Otherwise we still need the indirect branch, which we + * must restore after a potential direct branch write. + */ + br_diff -= 4; + if (in_range_b(br_diff)) { + i2 = B | (br_diff & 0x3fffffc); + i3 = NOP; + } else { + i2 = MTSPR | RS(TCG_REG_TB) | CTR; + i3 = BCCTR | BO_ALWAYS; + } + ppc64_replace4(jmp_rx, jmp_rw, i0, i1, i2, i3); } static void tcg_out_call_int(TCGContext *s, int lk, @@ -XXX,XX +XXX,XX @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, if (s->tb_jmp_insn_offset) { /* Direct jump. */ if (TCG_TARGET_REG_BITS == 64) { - /* Ensure the next insns are 8-byte aligned. */ - if ((uintptr_t)s->code_ptr & 7) { + /* Ensure the next insns are 8 or 16-byte aligned. */ + while ((uintptr_t)s->code_ptr & (have_isa_2_07 ? 15 : 7)) { tcg_out32(s, NOP); } s->tb_jmp_insn_offset[args[0]] = tcg_current_code_size(s); -- 2.34.1 The value previously chosen overlaps GUSA_MASK. Rename all DELAY_SLOT_* and GUSA_* defines to emphasize that they are included in TB_FLAGs. Add aliases for the FPSCR and SR bits that are included in TB_FLAGS, so that we don't accidentally reassign those bits. Fixes: 4da06fb3062 ("target/sh4: Implement prctl_unalign_sigbus") Resolves: https://gitlab.com/qemu-project/qemu/-/issues/856 Reviewed-by: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/sh4/cpu.h | 56 +++++++++++++------------ linux-user/sh4/signal.c | 6 +-- target/sh4/cpu.c | 6 +-- target/sh4/helper.c | 6 +-- target/sh4/translate.c | 90 ++++++++++++++++++++++------------------- 5 files changed, 88 insertions(+), 76 deletions(-) diff --git a/target/sh4/cpu.h b/target/sh4/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/cpu.h +++ b/target/sh4/cpu.h @@ -XXX,XX +XXX,XX @@ #define FPSCR_RM_NEAREST (0 << 0) #define FPSCR_RM_ZERO (1 << 0) -#define DELAY_SLOT_MASK 0x7 -#define DELAY_SLOT (1 << 0) -#define DELAY_SLOT_CONDITIONAL (1 << 1) -#define DELAY_SLOT_RTE (1 << 2) +#define TB_FLAG_DELAY_SLOT (1 << 0) +#define TB_FLAG_DELAY_SLOT_COND (1 << 1) +#define TB_FLAG_DELAY_SLOT_RTE (1 << 2) +#define TB_FLAG_PENDING_MOVCA (1 << 3) +#define TB_FLAG_GUSA_SHIFT 4 /* [11:4] */ +#define TB_FLAG_GUSA_EXCLUSIVE (1 << 12) +#define TB_FLAG_UNALIGN (1 << 13) +#define TB_FLAG_SR_FD (1 << SR_FD) /* 15 */ +#define TB_FLAG_FPSCR_PR FPSCR_PR /* 19 */ +#define TB_FLAG_FPSCR_SZ FPSCR_SZ /* 20 */ +#define TB_FLAG_FPSCR_FR FPSCR_FR /* 21 */ +#define TB_FLAG_SR_RB (1 << SR_RB) /* 29 */ +#define TB_FLAG_SR_MD (1 << SR_MD) /* 30 */ -#define TB_FLAG_PENDING_MOVCA (1 << 3) -#define TB_FLAG_UNALIGN (1 << 4) - -#define GUSA_SHIFT 4 -#ifdef CONFIG_USER_ONLY -#define GUSA_EXCLUSIVE (1 << 12) -#define GUSA_MASK ((0xff << GUSA_SHIFT) | GUSA_EXCLUSIVE) -#else -/* Provide dummy versions of the above to allow tests against tbflags - to be elided while avoiding ifdefs. */ -#define GUSA_EXCLUSIVE 0 -#define GUSA_MASK 0 -#endif - -#define TB_FLAG_ENVFLAGS_MASK (DELAY_SLOT_MASK | GUSA_MASK) +#define TB_FLAG_DELAY_SLOT_MASK (TB_FLAG_DELAY_SLOT | \ + TB_FLAG_DELAY_SLOT_COND | \ + TB_FLAG_DELAY_SLOT_RTE) +#define TB_FLAG_GUSA_MASK ((0xff << TB_FLAG_GUSA_SHIFT) | \ + TB_FLAG_GUSA_EXCLUSIVE) +#define TB_FLAG_FPSCR_MASK (TB_FLAG_FPSCR_PR | \ + TB_FLAG_FPSCR_SZ | \ + TB_FLAG_FPSCR_FR) +#define TB_FLAG_SR_MASK (TB_FLAG_SR_FD | \ + TB_FLAG_SR_RB | \ + TB_FLAG_SR_MD) +#define TB_FLAG_ENVFLAGS_MASK (TB_FLAG_DELAY_SLOT_MASK | \ + TB_FLAG_GUSA_MASK) typedef struct tlb_t { uint32_t vpn; /* virtual page number */ @@ -XXX,XX +XXX,XX @@ static inline int cpu_mmu_index (CPUSH4State *env, bool ifetch) { /* The instruction in a RTE delay slot is fetched in privileged mode, but executed in user mode. */ - if (ifetch && (env->flags & DELAY_SLOT_RTE)) { + if (ifetch && (env->flags & TB_FLAG_DELAY_SLOT_RTE)) { return 0; } else { return (env->sr & (1u << SR_MD)) == 0 ? 1 : 0; @@ -XXX,XX +XXX,XX @@ static inline void cpu_get_tb_cpu_state(CPUSH4State *env, target_ulong *pc, { *pc = env->pc; /* For a gUSA region, notice the end of the region. */ - *cs_base = env->flags & GUSA_MASK ? env->gregs[0] : 0; - *flags = env->flags /* TB_FLAG_ENVFLAGS_MASK: bits 0-2, 4-12 */ - | (env->fpscr & (FPSCR_FR | FPSCR_SZ | FPSCR_PR)) /* Bits 19-21 */ - | (env->sr & ((1u << SR_MD) | (1u << SR_RB))) /* Bits 29-30 */ - | (env->sr & (1u << SR_FD)) /* Bit 15 */ + *cs_base = env->flags & TB_FLAG_GUSA_MASK ? env->gregs[0] : 0; + *flags = env->flags + | (env->fpscr & TB_FLAG_FPSCR_MASK) + | (env->sr & TB_FLAG_SR_MASK) | (env->movcal_backup ? TB_FLAG_PENDING_MOVCA : 0); /* Bit 3 */ #ifdef CONFIG_USER_ONLY *flags |= TB_FLAG_UNALIGN * !env_cpu(env)->prctl_unalign_sigbus; diff --git a/linux-user/sh4/signal.c b/linux-user/sh4/signal.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/sh4/signal.c +++ b/linux-user/sh4/signal.c @@ -XXX,XX +XXX,XX @@ static void restore_sigcontext(CPUSH4State *regs, struct target_sigcontext *sc) __get_user(regs->fpul, &sc->sc_fpul); regs->tra = -1; /* disable syscall checks */ - regs->flags &= ~(DELAY_SLOT_MASK | GUSA_MASK); + regs->flags = 0; } void setup_frame(int sig, struct target_sigaction *ka, @@ -XXX,XX +XXX,XX @@ void setup_frame(int sig, struct target_sigaction *ka, regs->gregs[5] = 0; regs->gregs[6] = frame_addr += offsetof(typeof(*frame), sc); regs->pc = (unsigned long) ka->_sa_handler; - regs->flags &= ~(DELAY_SLOT_MASK | GUSA_MASK); + regs->flags &= ~(TB_FLAG_DELAY_SLOT_MASK | TB_FLAG_GUSA_MASK); unlock_user_struct(frame, frame_addr, 1); return; @@ -XXX,XX +XXX,XX @@ void setup_rt_frame(int sig, struct target_sigaction *ka, regs->gregs[5] = frame_addr + offsetof(typeof(*frame), info); regs->gregs[6] = frame_addr + offsetof(typeof(*frame), uc); regs->pc = (unsigned long) ka->_sa_handler; - regs->flags &= ~(DELAY_SLOT_MASK | GUSA_MASK); + regs->flags &= ~(TB_FLAG_DELAY_SLOT_MASK | TB_FLAG_GUSA_MASK); unlock_user_struct(frame, frame_addr, 1); return; diff --git a/target/sh4/cpu.c b/target/sh4/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/cpu.c +++ b/target/sh4/cpu.c @@ -XXX,XX +XXX,XX @@ static void superh_cpu_synchronize_from_tb(CPUState *cs, SuperHCPU *cpu = SUPERH_CPU(cs); cpu->env.pc = tb_pc(tb); - cpu->env.flags = tb->flags & TB_FLAG_ENVFLAGS_MASK; + cpu->env.flags = tb->flags; } #ifndef CONFIG_USER_ONLY @@ -XXX,XX +XXX,XX @@ static bool superh_io_recompile_replay_branch(CPUState *cs, SuperHCPU *cpu = SUPERH_CPU(cs); CPUSH4State *env = &cpu->env; - if ((env->flags & ((DELAY_SLOT | DELAY_SLOT_CONDITIONAL))) != 0 + if ((env->flags & (TB_FLAG_DELAY_SLOT | TB_FLAG_DELAY_SLOT_COND)) && env->pc != tb_pc(tb)) { env->pc -= 2; - env->flags &= ~(DELAY_SLOT | DELAY_SLOT_CONDITIONAL); + env->flags &= ~(TB_FLAG_DELAY_SLOT | TB_FLAG_DELAY_SLOT_COND); return true; } return false; diff --git a/target/sh4/helper.c b/target/sh4/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/helper.c +++ b/target/sh4/helper.c @@ -XXX,XX +XXX,XX @@ void superh_cpu_do_interrupt(CPUState *cs) env->sr |= (1u << SR_BL) | (1u << SR_MD) | (1u << SR_RB); env->lock_addr = -1; - if (env->flags & DELAY_SLOT_MASK) { + if (env->flags & TB_FLAG_DELAY_SLOT_MASK) { /* Branch instruction should be executed again before delay slot. */ env->spc -= 2; /* Clear flags for exception/interrupt routine. */ - env->flags &= ~DELAY_SLOT_MASK; + env->flags &= ~TB_FLAG_DELAY_SLOT_MASK; } if (do_exp) { @@ -XXX,XX +XXX,XX @@ bool superh_cpu_exec_interrupt(CPUState *cs, int interrupt_request) CPUSH4State *env = &cpu->env; /* Delay slots are indivisible, ignore interrupts */ - if (env->flags & DELAY_SLOT_MASK) { + if (env->flags & TB_FLAG_DELAY_SLOT_MASK) { return false; } else { superh_cpu_do_interrupt(cs); diff --git a/target/sh4/translate.c b/target/sh4/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/translate.c +++ b/target/sh4/translate.c @@ -XXX,XX +XXX,XX @@ void superh_cpu_dump_state(CPUState *cs, FILE *f, int flags) i, env->gregs[i], i + 1, env->gregs[i + 1], i + 2, env->gregs[i + 2], i + 3, env->gregs[i + 3]); } - if (env->flags & DELAY_SLOT) { + if (env->flags & TB_FLAG_DELAY_SLOT) { qemu_printf("in delay slot (delayed_pc=0x%08x)\n", env->delayed_pc); - } else if (env->flags & DELAY_SLOT_CONDITIONAL) { + } else if (env->flags & TB_FLAG_DELAY_SLOT_COND) { qemu_printf("in conditional delay slot (delayed_pc=0x%08x)\n", env->delayed_pc); - } else if (env->flags & DELAY_SLOT_RTE) { + } else if (env->flags & TB_FLAG_DELAY_SLOT_RTE) { qemu_fprintf(f, "in rte delay slot (delayed_pc=0x%08x)\n", env->delayed_pc); } @@ -XXX,XX +XXX,XX @@ static inline void gen_save_cpu_state(DisasContext *ctx, bool save_pc) static inline bool use_exit_tb(DisasContext *ctx) { - return (ctx->tbflags & GUSA_EXCLUSIVE) != 0; + return (ctx->tbflags & TB_FLAG_GUSA_EXCLUSIVE) != 0; } static bool use_goto_tb(DisasContext *ctx, target_ulong dest) @@ -XXX,XX +XXX,XX @@ static void gen_conditional_jump(DisasContext *ctx, target_ulong dest, TCGLabel *l1 = gen_new_label(); TCGCond cond_not_taken = jump_if_true ? TCG_COND_EQ : TCG_COND_NE; - if (ctx->tbflags & GUSA_EXCLUSIVE) { + if (ctx->tbflags & TB_FLAG_GUSA_EXCLUSIVE) { /* When in an exclusive region, we must continue to the end. Therefore, exit the region on a taken branch, but otherwise fall through to the next instruction. */ tcg_gen_brcondi_i32(cond_not_taken, cpu_sr_t, 0, l1); - tcg_gen_movi_i32(cpu_flags, ctx->envflags & ~GUSA_MASK); + tcg_gen_movi_i32(cpu_flags, ctx->envflags & ~TB_FLAG_GUSA_MASK); /* Note that this won't actually use a goto_tb opcode because we disallow it in use_goto_tb, but it handles exit + singlestep. */ gen_goto_tb(ctx, 0, dest); @@ -XXX,XX +XXX,XX @@ static void gen_delayed_conditional_jump(DisasContext * ctx) tcg_gen_mov_i32(ds, cpu_delayed_cond); tcg_gen_discard_i32(cpu_delayed_cond); - if (ctx->tbflags & GUSA_EXCLUSIVE) { + if (ctx->tbflags & TB_FLAG_GUSA_EXCLUSIVE) { /* When in an exclusive region, we must continue to the end. Therefore, exit the region on a taken branch, but otherwise fall through to the next instruction. */ tcg_gen_brcondi_i32(TCG_COND_EQ, ds, 0, l1); /* Leave the gUSA region. */ - tcg_gen_movi_i32(cpu_flags, ctx->envflags & ~GUSA_MASK); + tcg_gen_movi_i32(cpu_flags, ctx->envflags & ~TB_FLAG_GUSA_MASK); gen_jump(ctx); gen_set_label(l1); @@ -XXX,XX +XXX,XX @@ static inline void gen_store_fpr64(DisasContext *ctx, TCGv_i64 t, int reg) #define XHACK(x) ((((x) & 1 ) << 4) | ((x) & 0xe)) #define CHECK_NOT_DELAY_SLOT \ - if (ctx->envflags & DELAY_SLOT_MASK) { \ - goto do_illegal_slot; \ + if (ctx->envflags & TB_FLAG_DELAY_SLOT_MASK) { \ + goto do_illegal_slot; \ } #define CHECK_PRIVILEGED \ @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) case 0x000b: /* rts */ CHECK_NOT_DELAY_SLOT tcg_gen_mov_i32(cpu_delayed_pc, cpu_pr); - ctx->envflags |= DELAY_SLOT; + ctx->envflags |= TB_FLAG_DELAY_SLOT; ctx->delayed_pc = (uint32_t) - 1; return; case 0x0028: /* clrmac */ @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) CHECK_NOT_DELAY_SLOT gen_write_sr(cpu_ssr); tcg_gen_mov_i32(cpu_delayed_pc, cpu_spc); - ctx->envflags |= DELAY_SLOT_RTE; + ctx->envflags |= TB_FLAG_DELAY_SLOT_RTE; ctx->delayed_pc = (uint32_t) - 1; ctx->base.is_jmp = DISAS_STOP; return; @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) return; case 0xe000: /* mov #imm,Rn */ #ifdef CONFIG_USER_ONLY - /* Detect the start of a gUSA region. If so, update envflags - and end the TB. This will allow us to see the end of the - region (stored in R0) in the next TB. */ + /* + * Detect the start of a gUSA region (mov #-n, r15). + * If so, update envflags and end the TB. This will allow us + * to see the end of the region (stored in R0) in the next TB. + */ if (B11_8 == 15 && B7_0s < 0 && (tb_cflags(ctx->base.tb) & CF_PARALLEL)) { - ctx->envflags = deposit32(ctx->envflags, GUSA_SHIFT, 8, B7_0s); + ctx->envflags = + deposit32(ctx->envflags, TB_FLAG_GUSA_SHIFT, 8, B7_0s); ctx->base.is_jmp = DISAS_STOP; } #endif @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) case 0xa000: /* bra disp */ CHECK_NOT_DELAY_SLOT ctx->delayed_pc = ctx->base.pc_next + 4 + B11_0s * 2; - ctx->envflags |= DELAY_SLOT; + ctx->envflags |= TB_FLAG_DELAY_SLOT; return; case 0xb000: /* bsr disp */ CHECK_NOT_DELAY_SLOT tcg_gen_movi_i32(cpu_pr, ctx->base.pc_next + 4); ctx->delayed_pc = ctx->base.pc_next + 4 + B11_0s * 2; - ctx->envflags |= DELAY_SLOT; + ctx->envflags |= TB_FLAG_DELAY_SLOT; return; } @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) CHECK_NOT_DELAY_SLOT tcg_gen_xori_i32(cpu_delayed_cond, cpu_sr_t, 1); ctx->delayed_pc = ctx->base.pc_next + 4 + B7_0s * 2; - ctx->envflags |= DELAY_SLOT_CONDITIONAL; + ctx->envflags |= TB_FLAG_DELAY_SLOT_COND; return; case 0x8900: /* bt label */ CHECK_NOT_DELAY_SLOT @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) CHECK_NOT_DELAY_SLOT tcg_gen_mov_i32(cpu_delayed_cond, cpu_sr_t); ctx->delayed_pc = ctx->base.pc_next + 4 + B7_0s * 2; - ctx->envflags |= DELAY_SLOT_CONDITIONAL; + ctx->envflags |= TB_FLAG_DELAY_SLOT_COND; return; case 0x8800: /* cmp/eq #imm,R0 */ tcg_gen_setcondi_i32(TCG_COND_EQ, cpu_sr_t, REG(0), B7_0s); @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) case 0x0023: /* braf Rn */ CHECK_NOT_DELAY_SLOT tcg_gen_addi_i32(cpu_delayed_pc, REG(B11_8), ctx->base.pc_next + 4); - ctx->envflags |= DELAY_SLOT; + ctx->envflags |= TB_FLAG_DELAY_SLOT; ctx->delayed_pc = (uint32_t) - 1; return; case 0x0003: /* bsrf Rn */ CHECK_NOT_DELAY_SLOT tcg_gen_movi_i32(cpu_pr, ctx->base.pc_next + 4); tcg_gen_add_i32(cpu_delayed_pc, REG(B11_8), cpu_pr); - ctx->envflags |= DELAY_SLOT; + ctx->envflags |= TB_FLAG_DELAY_SLOT; ctx->delayed_pc = (uint32_t) - 1; return; case 0x4015: /* cmp/pl Rn */ @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) case 0x402b: /* jmp @Rn */ CHECK_NOT_DELAY_SLOT tcg_gen_mov_i32(cpu_delayed_pc, REG(B11_8)); - ctx->envflags |= DELAY_SLOT; + ctx->envflags |= TB_FLAG_DELAY_SLOT; ctx->delayed_pc = (uint32_t) - 1; return; case 0x400b: /* jsr @Rn */ CHECK_NOT_DELAY_SLOT tcg_gen_movi_i32(cpu_pr, ctx->base.pc_next + 4); tcg_gen_mov_i32(cpu_delayed_pc, REG(B11_8)); - ctx->envflags |= DELAY_SLOT; + ctx->envflags |= TB_FLAG_DELAY_SLOT; ctx->delayed_pc = (uint32_t) - 1; return; case 0x400e: /* ldc Rm,SR */ @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) fflush(stderr); #endif do_illegal: - if (ctx->envflags & DELAY_SLOT_MASK) { + if (ctx->envflags & TB_FLAG_DELAY_SLOT_MASK) { do_illegal_slot: gen_save_cpu_state(ctx, true); gen_helper_raise_slot_illegal_instruction(cpu_env); @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) do_fpu_disabled: gen_save_cpu_state(ctx, true); - if (ctx->envflags & DELAY_SLOT_MASK) { + if (ctx->envflags & TB_FLAG_DELAY_SLOT_MASK) { gen_helper_raise_slot_fpu_disable(cpu_env); } else { gen_helper_raise_fpu_disable(cpu_env); @@ -XXX,XX +XXX,XX @@ static void decode_opc(DisasContext * ctx) _decode_opc(ctx); - if (old_flags & DELAY_SLOT_MASK) { + if (old_flags & TB_FLAG_DELAY_SLOT_MASK) { /* go out of the delay slot */ - ctx->envflags &= ~DELAY_SLOT_MASK; + ctx->envflags &= ~TB_FLAG_DELAY_SLOT_MASK; /* When in an exclusive region, we must continue to the end for conditional branches. */ - if (ctx->tbflags & GUSA_EXCLUSIVE - && old_flags & DELAY_SLOT_CONDITIONAL) { + if (ctx->tbflags & TB_FLAG_GUSA_EXCLUSIVE + && old_flags & TB_FLAG_DELAY_SLOT_COND) { gen_delayed_conditional_jump(ctx); return; } /* Otherwise this is probably an invalid gUSA region. Drop the GUSA bits so the next TB doesn't see them. */ - ctx->envflags &= ~GUSA_MASK; + ctx->envflags &= ~TB_FLAG_GUSA_MASK; tcg_gen_movi_i32(cpu_flags, ctx->envflags); - if (old_flags & DELAY_SLOT_CONDITIONAL) { + if (old_flags & TB_FLAG_DELAY_SLOT_COND) { gen_delayed_conditional_jump(ctx); } else { gen_jump(ctx); @@ -XXX,XX +XXX,XX @@ static void decode_gusa(DisasContext *ctx, CPUSH4State *env) } /* The entire region has been translated. */ - ctx->envflags &= ~GUSA_MASK; + ctx->envflags &= ~TB_FLAG_GUSA_MASK; ctx->base.pc_next = pc_end; ctx->base.num_insns += max_insns - 1; return; @@ -XXX,XX +XXX,XX @@ static void decode_gusa(DisasContext *ctx, CPUSH4State *env) /* Restart with the EXCLUSIVE bit set, within a TB run via cpu_exec_step_atomic holding the exclusive lock. */ - ctx->envflags |= GUSA_EXCLUSIVE; + ctx->envflags |= TB_FLAG_GUSA_EXCLUSIVE; gen_save_cpu_state(ctx, false); gen_helper_exclusive(cpu_env); ctx->base.is_jmp = DISAS_NORETURN; @@ -XXX,XX +XXX,XX @@ static void sh4_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs) (tbflags & (1 << SR_RB))) * 0x10; ctx->fbank = tbflags & FPSCR_FR ? 0x10 : 0; - if (tbflags & GUSA_MASK) { +#ifdef CONFIG_USER_ONLY + if (tbflags & TB_FLAG_GUSA_MASK) { + /* In gUSA exclusive region. */ uint32_t pc = ctx->base.pc_next; uint32_t pc_end = ctx->base.tb->cs_base; - int backup = sextract32(ctx->tbflags, GUSA_SHIFT, 8); + int backup = sextract32(ctx->tbflags, TB_FLAG_GUSA_SHIFT, 8); int max_insns = (pc_end - pc) / 2; if (pc != pc_end + backup || max_insns < 2) { /* This is a malformed gUSA region. Don't do anything special, since the interpreter is likely to get confused. */ - ctx->envflags &= ~GUSA_MASK; - } else if (tbflags & GUSA_EXCLUSIVE) { + ctx->envflags &= ~TB_FLAG_GUSA_MASK; + } else if (tbflags & TB_FLAG_GUSA_EXCLUSIVE) { /* Regardless of single-stepping or the end of the page, we must complete execution of the gUSA region while holding the exclusive lock. */ @@ -XXX,XX +XXX,XX @@ static void sh4_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs) return; } } +#endif /* Since the ISA is fixed-width, we can bound by the number of instructions remaining on the page. */ @@ -XXX,XX +XXX,XX @@ static void sh4_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs) DisasContext *ctx = container_of(dcbase, DisasContext, base); #ifdef CONFIG_USER_ONLY - if (unlikely(ctx->envflags & GUSA_MASK) - && !(ctx->envflags & GUSA_EXCLUSIVE)) { + if (unlikely(ctx->envflags & TB_FLAG_GUSA_MASK) + && !(ctx->envflags & TB_FLAG_GUSA_EXCLUSIVE)) { /* We're in an gUSA region, and we have not already fallen back on using an exclusive region. Attempt to parse the region into a single supported atomic operation. Failure @@ -XXX,XX +XXX,XX @@ static void sh4_tr_tb_stop(DisasContextBase *dcbase, CPUState *cs) { DisasContext *ctx = container_of(dcbase, DisasContext, base); - if (ctx->tbflags & GUSA_EXCLUSIVE) { + if (ctx->tbflags & TB_FLAG_GUSA_EXCLUSIVE) { /* Ending the region of exclusivity. Clear the bits. */ - ctx->envflags &= ~GUSA_MASK; + ctx->envflags &= ~TB_FLAG_GUSA_MASK; } switch (ctx->base.is_jmp) { -- 2.34.1